Metal Partitioning Among Tissues and Exoskeleton of Palaemonetes Pugio and Its Role in Depuration and Trophic Transfer

Home , Arthropod exoskeleton, Palaemonetes pugio

Louisiana State University LSU Digital Commons

LSU Historical Dissertations and Theses Graduate School

2001 Metal Partitioning Among Tissues and Exoskeleton of Palaemonetes Pugio and Its Role in Depuration and Trophic Transfer. Kristen Annette Keteles Louisiana State University and Agricultural & Mechanical College

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_disstheses

Recommended Citation Keteles, Kristen Annette, "Metal Partitioning Among Tissues and Exoskeleton of Palaemonetes Pugio and Its Role in Depuration and Trophic Transfer." (2001). LSU Historical Dissertations and Theses. 380. https://digitalcommons.lsu.edu/gradschool_disstheses/380

This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please contact [email protected]. U MI MICROFILMED 2002

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

ProQuest Information and Learning 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA 800-521-0600

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. METAL PARTITIONING AMONG TISSUES AND EXOSKELETON OF PALAEMONETES PUGIO AND ITS ROLE IN DEPURATION AND TROPHIC TRANSFER

A Dissertation

Submitted to the Graduate Faculty o f the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree o f Doctor o f Philosophy

The Department o f Biological Sciences

by Kristen A. Keteles B.S. Coastal Carolina University 1995 December 2001

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 3031762

__ ® UMI

UMI Microform 3031762 Copyright 2002 by Bell & Howell Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.

Bell & Howell Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGMENTS

This dissertation is dedicated to my parents, especially my father who encouraged my

interests in biology. I would like to express profound appreciation for Dr. John Fleeger

for his advice and support during the production o f this dissertation. This work would not

be possible without the help of Dr. Robert Gambrell and Medhi Aarabi with the metal

analyses. I also thank my minor professor. Dr. Ralph Portier. and the other members o f

my committee. Dr. Kevin Carman. Dr. W illiam B. Stickle. Dr. James Catallo, and Dr.

Robert Gambrell for advice. Statistical advice was provided by B. Marx. I also thank Dr.

G. Tita. Dr. N. Atilla. G. Puckett. S. Hymel, and A. Martin for help with various aspects

o f this dissertation research. I also want to thank my teammates from LSU cycling for

their support. This study was funded by the Office o f Naval Research grant # N00014-

99-1-0023 and a Sigma Xi Grant-in-Aid o f Research.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS

ACKNOWLEDGMENTS...... ii

ABSTRACT...... iv

CHAPTER 1 GENERAL INTRODUCTION...... 1

2 THE CONTRIBUTION OF ECDYSIS TO THE FATE OF COPPER, ZINC, AND CADMIUM IN GRASS SHRIMP, PALAEMONETES PUGIO HOLTHIUS...... 13 Introduction ...... 14 Methods...... 16 Results...... 19 Discussion...... 24

3 METAL UPTAKE AND PARTITIONING AMONG TISSUES AND EXOSKELF.TON OF THE GRASS SHRIMP. PALAEMONETES PUGIO 29 Introduction ...... 30 Methods...... 33 Results...... 37 Discussion...... 49

4 TROPHIC TRANSFER OF CU. ZN. AND CD ASSOCIATED WITH EXOSKELETON AND TISSUES OF THE GRASS SHRIMP. PALAEMONETES PUGIO. TO A PREDATORY FISH. FUNDULUS GMNDIS...... 58 Introduction ...... 59 Methods...... 61 Results...... 64 Discussion...... 78

5 SUMMARY AND CONCLUSIONS...... 85

LITERATURE CITED...... 97

APPENDIX: LETTER OF PERMISSION...... 103

VITA ...... 104

iii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT

The objective o f this dissertation was to determine the partitioning o f Cu, Zn and

Cd between tissues and exoskeleton o f the grass shrimp,Palaemonetes pugio Holthius.

and to examine the consequences o f this association. The role o f the exoskeleton in the

potential for depuration o f metals through ecdysis was examined first. The portion

eliminated with the exuviae varied for each metal; 11% Cu, 18% Zn. and 26% Cd o f the

total intermolt body burden was associated with exuviae. Some fraction o f Cu and Cd in

the exoskeleton was likely reabsorbed before molting but excess Zn was likely depurated

through excretion. The influence o f salinity and source o f exposure on metal partitioning

among tissue and exoskeleton was then investigated. Grass shrimp were exposed to

aqueous metals in 5.18. and 30 %o seawater or fed metal-enriched algal epiphytes from

the stemso f Spartina alterniflora. Surface-adsorbed metals ranged from 0-20 % o f the

whole-body burden with the lowest values at 5 %o. Metals associated with the

exoskeleton matrix ranged from 7-45% with highest values for Zn and Cd. Tissue Cu and

Zn burden showed very little variability among the salinity and dietary-exposure

treatments likely due to physiological regulation. Finally, the role o f partitioning among

tissue and exoskeleton in the trophic transfer o f metals to the predatory' fishFundulus

grandis was examined. F. grandis were fed tissues, total exoskeleton (including the

adsorbed metals), or exoskeleton matrix fractions o f P. pugio separately for 10 d.

Assimilation efficiencies o f Cu and Cd associated with the exoskeleton matrix and tissues

o fP. pugio by F. grandis were relatively low and did not differ. Zn associated with the

tissues was highly available, but exoskeleton matrix-bound Zn was essentially

unavailable. Surface-adsorbed Cd and Zn were highly available to F. grandis. Surface-

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. adsorbed metals should be considered in future studies because they can be an important

vector in food web transfer o f metals. However, metals adsorbed to the epicuticle likely

do not interact with receptor sites or contribute to toxicity and should be taken into

account when determining toxicological endpoints.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 1

GENERAL INTRODUCTION

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Crustaceans are important members o f marine communities ranging from

estuaries to the deep sea. Planktonic and benthic crustaceans are abundant, high-biomass

invertebrates that exhibit high population consumption rates and account for a significant

fraction o f the flow o f energy through all marine environments. Crustaceans, including

copepods. amphipods. mysids. euphausiids, various decapods and many others, serve as

principal prey for a large number o f higher-level predators. Benthic and pelagic

crustaceans have also been shown to have high contaminant body burdens o f trace metals

(Depledgeet al. 1993; Watras and Bloom 1992) that may cause significant toxic effects

and enhance the potential for the transfer o f contaminants to higher trophic levels

(Russell el al. 1999). Crustaceans are also frequently used as sentinel species at field

monitoring sites (Petersonet al. 1996) and as model species in laboratory toxicological

studies (Fisher and Foss 1993; Key and Fulton 1993; Klerks 1999).

Due to urban and agricultural runoff and the release o f industrial and municipal

waste, estuarine crustaceans are exposed to the impacts o f trace-metal pollution including

copper, zinc, and cadmium. Cu plays a key role in growth and in the oxygen carrying

pigment, hemocyanin. (Hebeiet al. 1997). and Zn is an essential cofactor in enzyme

function in crustaceans (Bryan 1968). However, excess Cu and Zn are toxic, and

sublethal concentrations may lead to metabolic and cellular disturbances ( Hebei et al.

1997; Depledge el al. 1993; Bryan 1968). Because Cu and Zn are essential to metabolic

function, they are subject to physiological regulation in crustaceans. On the other hand.

Cd has no known physiological function, and Cd at low concentrations can induce toxic

effects (Nimmoet al. 1977; Sprague 1986). Furthermore, Cd is chemically similar to Ca

and can displace Ca uptake in crustaceans (Guerin and Stickle 1995; Wallace and Lopez

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1997; Zanders and Rojas 1996). Toxic responses caused by Cd include tissue damage,

abnormal development, reduced growth, and reduced fecundity (Nimmo et al. 1977).

Although crustaceans are highly variable in body size, habitat and ecological role,

they share the presence o f a chitinous exoskeleton that potentially allows partitioning o f

trace metals between the exoskeleton and soft tissues. Chitin, a polysaccharide composed

o f amino sugars, is a principal constituent o f the inner procuticle layer o f crustacean

exoskeletons (Stevenson 1985) Cations bind to chitin by forming ionic bonds with

nitrogen side groups within the chitin fibrils o f the procuticle and epicuticle Relatively

little is known o f the role that partitioning with the exoskeleton plays in the

bioaccumulation and fate o f metals in crustaceans. In addition, the exoskeleton is subject

to periodic molting and associated cycles o f calcium deposition and release. Crustaceans

harden the procuticle and epicuticle following molting by the deposition o f calcium salts.

Ca release from the exoskeleton occurs prior to molting, and Ca is stored in cells

associated with the hepatopancreas while molting occurs (Zhuang and Aheam 1996).

Therefore, the distribution o f metal cations, chemically similar to Ca (e.g., Zn and Cd).

among tissues and exoskeleton may vary throughout the molt cycle in a fashion similar to

Ca. It is likely that deleterious cations become bound in this matrix o f the exoskeleton in

association with periodic Ca deposition. Some fraction o f these cationic metals, however,

may be subject to mobilization and release to the tissues along with Ca prior to molting.

Other metals become associated with the exoskeleton through different mechanisms, but

may also be subject to physiological variation associated the molt cycle. For example,

Cu-containing proteins similar to the oxygen-carrier pigment hemocyanin are involved in

cross-linking and hardening o f the exoskeleton after molting (Terwilliger 1999). Thus.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. metal distribution among tissues and the matrix o f the exoskeleton probably varies with

respect to metal species and crustacean physiological condition.

Many organisms precipitate metals in metal-rich granules to store and/or excrete

essential and toxic metals (Brown 1982). Metal granules are insoluble, and therefore,

unavailable to predators (Wang et al. 1999; Wallace and Lopez 1997). These mineral

deposits are found both intracellularly and extracellularly and accrete Zn. Cu, Ca, Cd. Fe.

and Pb. In crustaceans, these granules have been found in the hepatopancreas. midgut.

digestive gland, malpighian tubules, and kidney and serve in the role o f excretion (Brown

1982). However, localization in granules can vary with the molt cycle, and Cu-containing

granules in the hepatopancreas are released during molting and influence

compartmentalization within the tissues (Brown 1982; Bryan 1968). Following ecdysis.

calcium-containing granules are used in exoskeleton construction (Greenaway and

Farrelly 1991). and crustaceans may also sequester metals in granule form in the

exoskeleton for detoxification or elimination with the molt. However, to my knowledge

there are no reports that metals other than calcium are precipitated in granules in the

exoskeleton.

Investigators frequently assume that crustaceans routinely depurate metals by

molting (Bertine and Goldberg 1972: Khan et al. 1989; Reinfelder and Fisher 1994).

Such generalizations may not always be warranted, and the fate o f metals associated with

molting appears to depend, at least partially, on the route o f exposure to metals. Fowler

et al. (1971) found that Zn accumulated from the dissolved phase associated with

interstitial spaces in the exoskeleton and that up to 41% o f the Zn body burden was lost

with the molted exoskeleton. Fowlerel al. (1970) found that Zn accumulated from food.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. however, became localized on ligand-binding sites on the underside o f the exoskeleton,

and in long-term experiments, only 1% accumulated from food was eliminated with each

molt. Much less is known about the importance o f molting to the fate o f Cu and Cd in

crustaceans.

In addition to deposition within the chitin matrix, metals may directly bind to the

outer surface o f the epicuticle by adsorptive processes. Chitin and associated proteins

have many hydroxyl, nitrogen, and sulfur-containing groups that may serve as ligand-

binding sites for cationic metals (Stevenson 1986). Many such sites are likely to be quite

common on the crustacean epicuticle and on the newly-exposed surfaces o f exuviae,

enhancing the potential for adsorption. Consequently, if adsorption o f metals occurs

directly to the epicuticle without association with biological tissues, adsorbed metals may

not contribute directly to toxic effects. Ingestion o f exuviae following molting is.

however, common in many crustaceans, leading to the possibility that adsorbed metals

may enter tissues by secondary processing after ingestion and absorption through the gut.

A limited number o f studies have examined the partitioning o f metals between

soft tissues and chitinous exoskeletons. Fowleret al. (1970) found that 30-66% o f the

b5Zn body burden was consistently contained in the exoskeletons o f a euphausiid and a

prawn. Metal concentrations have been reported to be higher in grass shrimp

exoskeletons than in muscle tissue (Khanet al. 1989). Munger and Hare (1997) reported

that 19% o f the cadmium body burden was associated with the exoskeleton o f a

cladoceran. Reinfelder and Fisher (1994) observed that copepod exoskeletons contained

97% o f their trace-metal burden. Uca rapax. a fiddler crab, was reported to accumulate

80% o f its Cd burden in the carapace (Zanders and Rojas 1996). O f these, only Munger

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and Hare (1997). Zanders and Rojas (1996) and Reinfelder and Fisher (1994)

distinguished between the surface-sorbed and the matrix-bound fractions o f the

exoskeleton; however, none o f the studies cited above were designed to quantify the

amount o f the metal burden that was superficially adsorbed to the epicuticle. Hare (1992)

acknowledged that metals can be associated with the surface o f aquatic insect

exoskeletons, and reported that 25% o f the Cu. 10% o f the Zn and up to 40% o f the Cd

body burdens were associated with the surface o f nymph Hexagenia limbata exoskeleton.

Current knowledge appears to be insufficient to generalize the relative contribution o f the

adsorbed fraction to the total body burden in crustaceans.

A possible explanation for the discrepancy in reported exoskeleton metal burden

is the influence o f salinity on the binding affinities o f metal cations (Libes 1992).

Because seawater has a high concentration o f ions, interactions between solutes occur that

can influence the effective concentration or activity o f an ion. The rate and degree at

which a dissolved metal adsorbs to clay minerals and organic matter, such as the

crustacean epicuticle. depends upon its elemental nature, concentrations o f other solutes,

and the abundance o f particulate matter (Libes 1992). Salinity influences bioavailability

(Guerin and Stickle 1995; Zanders and Rojas 1996) and may therefore influence the metal

deposition within the chitinous matrix and uptake by the tissues and may thereby alter

panitioning.

Partitioning among exoskeleton and tissue pools in crustaceans may also vary as a

function o f exposure route. Metals that enter with the ingestion o f food (and are

subsequently absorbed by the gut) or are absorbed via the gills are subject to

physiological regulating mechanisms. The fate o f such contaminants is to be stored in

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. various tissue and exoskeleton pools or ultimately metabolized or excreted (Fowleret al.

1971). Metals obtained from food w ill likely associate with the matrix o f the exoskeleton

only after absorption and transport through the circulator)' system. Thus, the route o f

exposure should be considered when examining contaminant interactions with the

exoskeleton.

Partitioning o f metals in the crustacean exoskeleton may also greatly influence

trophic transfer from crustacean prey to predators or to crustaceans that ingest their

exuviae. Some predators reject the crustacean exoskeleton, and a high assimilation

efficiency was observed for a predator feeding on metal-enriched crustacean prey perhaps

because the predator o f interest selectively ingested soft tissues (Smokorowskiet al.

1998). Reinfelder and Fisher (1994) found a strong correlation between a predatory

fish's absorption efficiency o f metals and the non-exoskeleton tissues o f copepod prey.

Chitin degrading enzymes are extremely rare among fish and crustacean predators, and it

is likely that metals bound in the matrix o f exoskeleton at the time o f ingestion are

relatively unavailable for absorption. Alternatively, metals adsorbed to the exoskeleton

may be readily available for absorption in the acidic digestive tract o f predators and o f

crustaceans that ingest their own exuviae. It is imperative to understand the mechanisms

and the partitioning o f metals in order to predict trophic transfer and food-web effects,

and ultimately the potential for human-health effects.

The overall objective o f this dissertation research was to determine the association

o f Cu. Zn and Cd with exoskeleton o f the grass shrimp. Palaemonetes pugio Hoithius.

including association through surface-adsorption. This study was designed to mimic

natural conditions as closely as possible, and stable forms o f Cu. Zn. and Cd were used

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. because the use o f radiotracers may produce artifacts due to differences in

biogeochemical behavior. For example, Paulson and Gendron (2001) found that the

association o f 64Cu with organic particles was 70% higher than that o f natural Cu. P

pugio was used as a model crustacean due to its ecological relevance as an abundant

resident o f estuaries and as an important member o f the salt-marsh food web (Kneib

1987). Furthermore. P.pugio is often exposed to elevated metals and can be a vector in

bioaccumulation by higher trophic levels (Smith and Weis 1997; Khan et al 1989;

Nimmo et al.. 1977). Another objective was to understand the role o f the exoskeleton in

the potential depuration o f metals through ecdysis. Additionally, the effect o f salinity and

source o f exposure on partitioning among tissue and exoskeleton was investigated.

Finally, the ecological significance o f the role o f partitioning among the exoskeleton and

tissues in the trophic transfer o f metals was addressed. The following null hypotheses

were tested in my dissertation:

• Ecdysis does not influence the Cu, Zn. and Cd body burdens in Palaemonetes pugio

• Salinity does not influence partitioning o f Cu. Zn. and Cd between tissue and exoskeleton (both surface-adsorbed and matrix-sequestered fractions) in P. pugio

• The route o f exposure does not influence partitioning o f Cu. Zn and Cd between tissue and exoskeleton (both surface-adsorbed and matrix- sequestered fractions) in P. pugio.

• Partitioning between tissue and exoskeleton (both surface-adsorbed and matrix- sequestered fractions) does not influence the trophic transfer o f Cu. Zn. and Cd to a fish predator (Fundulus grandis).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The specific objective o f the second chapter o f this dissertation was to quantify

the sequestration o f Cu, Zn, and Cd in sublethal concentrations from the aqueous phase

by the exoskeleton o f grass shrimp and to determine what fraction o f these trace metals

are depurated through ecdysis. Elimination o f metals in decapod crustaceans can be

attributed to excretion, flux across the body surfaces and gills (Bryan 1968) as well as to

molting. It has been suggested that the crustacean exoskeleton may sequester metals and

contribute to elimination (depuration) through ecdysis (Reinfelder and Fisher 1994;

Bertine and Goldberg 1972; Khan et al. 1989; Smokorowskiet al. 1998). My study was

designed to determine the role o f ecdysis in the fate o f Cu. Zn. and Cd in P. pugio.

Another topic addressed in Chapter 2 was the role o f surface adsorption on

exuviae shed in metal-enriched water Ligand-binding sites may be more common on the

underside o f the molted exoskeleton than on the exposed epicuticle. Therefore, after

molting, adsorption to newly exposed binding sites could contribute to an increase in the

concentrations o f metals in exuviae cast in contaminated water. White and Rainbow

(1984) found that exuviae o f Palaemon elegans take up Cu, Zn. and Cd from the ambient

medium, and it has been speculated that adsorption o f metals onto the shed molt maybe

an important phenomenon in the cycling o f trace metals in aquatic systems (Wang et al.

1996). Additionally, many crustaceans, particularly in densely-populated farming

conditions, ingest exuviae in order to compensate for the energetic loss o f exuvial

production (Guillaume 1997). Therefore, adsorption o f metals from the aqueous phase by

the exuviae was measured in the second chapter o f dissertation because o f its potential

importance in aquaculture. This second chapter has been accepted for publication in

Marine Pollution Bulletin.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The objective o f Chapter 3 was to determine how the source o f exposure (aqueous

or dietary) and salinity influences the uptake and partitioning o f Cu. Zn, and Cd among

the tissues and exoskeleton (both surface-adsorbed and matrix-bound) o f P. pugio. A

variety o f metals (i.e.. Cu, Zn, Cd. Co. Se. Hg) associated with exoskeletons have been

measured in diverse groups o f crustaceans and authors report widely varying results

(Zanders and Rojas 1996: Khan et al. 1989; Munger and Hare 1997; Depledge et al.

1993; Reinfelder and Fisher 1994; White and Rainbow 1984; Nimmo et al. 1977; Fowler

et al. 1970: Bryan 1968). The highest reported values (97% o f the total body Cd. Zn. and

Co burden) o f the exoskeleton are reported in marine planktonic copepods (Reinfelder

and Fisher 1994): however, a freshwater cladoceran was shown to have only 19% o f the

Cd burden in the exoskeleton (Munger and Hare 1997). Variability among these studies

could be due to differences in physiology, metal-species behavior, salinity, and source o f

exposure. Because salinity influences metal speciation. and subsequently metal solubility

and bioavailability (Guerin and Stickle 1995; Zanders and Rojas 1996). salinity effects on

uptake and partitioning from aqueous exposure were examined by aqueous exposure at

three salinities (5. 18. and 30%o).

In order to examine the effect o f dietary exposure on partitioning, metal-exposed

epiphytic algae associated with the stems o f Spartina alterniflora were used as the food

source because grass shrimp are know to ingest epiphytes (Morgan 1980) including those

associated with S. alterniflora (Fleegeret al. 1999). S. alterniflora is instrumental in

cycling trace metals in wetlands because metals are released through the salt glands in the

leaves (Burkeet al. 2000) and released metals may subsequently become available to

epiphytes. Catallo et al. (1996) demonstrated that endosymbiotic yeast in S. alterniflora

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. are capable o f assimilating Cu, Zn. Cd. and Ni, and perhaps other microbial symbionts

may also be important in the cycling o f trace metals in salt marshes through assimilation

and trophic transfer to herbivores such as P. pugio.

A major objective of Chapter 3 was to quantify surface-adsorbed and matrix-

bound metals under a range o f uptake and salinity conditions. Although many authors

(Reinfelder and Fisher 1994; Zanders and Rojas 1996; Smokerowski et al. 1998)

acknowledge the potential significance of adsorption, no studies have attempted to

directly quantify the distribution between metals adsorbed to the surface and metals

sequestered within the chitin matrix in crustaceans. A strong chelator (EDTA) was used

to remove surface-adsorbed metals in P. pugio. This process does not solubilize metals

sequestered within the procuticle and served as the basis to distinguish between surface-

adsorbed and matrix-sequestered metals.

The objective o f Chapter 4 was to determine the role o f the crustacean

exoskeleton in the trophic transfer o f metals from a crustacean(P. pugio) to a fish

predator (Fundulus grandis Baird and Girard). F. grandis is an important estuarine

resident, frequently used in toxicological studies (Smith and Weis 1997; Davis et al.

1998). F. grandis feeds on invertebrate prey including P. pugio (Kneib 1987).

Assimilation efficiencies o f metals by predators are dependent on the chemical species

and the tissues in which they are adsorbed (Chen et al. 2000), and therefore, partitioning

among tissues and exoskeleton of prey may influence bioavailability to higher trophic

levels. The assimilation efficiency o f metals (defined as the amount taken up by the

tissues (Penry 1998) adsorbed to the surface of the exoskeleton, incorporated into the

chitin matrix, and associated w ith the soft tissues were specifically measured and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. compared. Grass shrimp exoskeleton, with or without removal o f surface-adsorbed

metals, or tissues were fed toF. grandis in separate gelatin capsules for 10 days.

Previous authors have correlated assimilation efficiencies with the proportion o f metals in

soft tissues o f crustacean prey (Munger and Hare 1997; Reinfelder and Fisher; 1994. Ni et

al. 2000; Wang et al. 1999). However, direct measurements o f metal assimilation from

crustacean soft tissue or exoskeleton has not been reported. It has been suggested that

metals associated with the exoskeleton are unavailable to higher trophic levels due to the

indigestibly of chitin. but this presumption has not been directly tested. Furthermore, the

assimilation o f surface-adsorbed metals by higher trophic levels has not been determined.

Surface-adsorbed metals may be an important vector in food web transfer because metals

may be solublized in the gastric fluids o f predators. Finally, the significance o f this

dissertation is discussed in the concluding chapter (Chapter 5).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPER 2

THE CONTRIBUTION OF ECDYSIS TO THE FATE OF COPPER, ZINC AND CADMIUM IN GRASS SHRIMP, PALAEMONETES PUGIO HOLTHIUS*

*Reprinted from Marine Pollution Bulletin (in press) Keteles, K.A. and Fleeger. J. W .. The contribution o f ecdysis to the fate of copper, zinc and cadmium in grass shrimp Palaemonetes pugio Hothius, with permission from Elsevier Science.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. INTRODUCTION

There are several possible fates(e.g., excretion, sequestration by metal-binding

proteins) o f trace metals after bioaccumulation in aquatic organisms (Hare 1992). In

crustaceans, some fraction o f the vvhole-body burden o f trace metals is associated with the

chitinous exoskeleton (Munger and Hare 1997; Zanders and Rojas 1996; Reinfelder and

Fisher 1994; Depledgeel al. 1993: Weekset al. 1992; Khan et al. 1989; Nimmo et al.

1977; White and Rainbow 1984; Fowler et al. 1971: Bryan 1968). This relationship

appears to be highly variable among taxa and metal species, but the fraction associated

with the exoskeleton may represent a significant portion o f the total body burden (reports

cited above vary from 19-97%). To further complicate exoskeleton-metal associations,

metals may either adsorb to the surface o f the exoskeleton or bind to the inner

exoskeleton matrix after uptake and transport through the hemolymph. The adsorbed

fraction is rarely considered (Hare 1992). Also, as an important aspect o f crustacean

physiology, molting may intluence metal concentrations and the distribution between the

soft tissues and exoskeleton (Weekset al. 1992). It has been suggested that the

crustacean exoskeleton may sequester metals and contribute to elimination (depuration)

through ecdysis (Reinfelder and Fisher 1994; Bertine and Goldberg 1972; Khan el al.

1989: Smokorowskiel al. 1998).

The molt cycle, and its associated storage and deposition o f calcium salts, may

facilitate the sequestration of trace metals within the crustacean exoskeleton. Crustaceans

harden the procuticle and epicuticle following molting by the deposition o f calcium salts.

Ca release from the exoskeleton occurs prior to molting (Stevenson 1985). and Ca in

some crustaceans is stored in cells associated with the hepatopancreas and in the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. gastrof iths o f the proventricuJus while molting occurs (Adams et al. 1982; Zhuang and

Aheam 1996). Metal cations (e.g.. zinc and cadmium) bind to chitin by forming ionic

bonds with nitrogen side groups within the chitin fibrils o f the procuticle. It is likely that

cations become bound in this matrix o f the exoskeleton in association with Ca deposition.

Some fraction o f these cationic metals, however, may be subject to mobilization and

release to the tissues along with Ca prior to molting. Other metals become associated

with the exoskeleton through different mechanisms, but may also be subject to

physiological variation associated the molt cycle. For example, copper-containing

proteins similar to the oxygen-carrier pigment hemocyanin are involved in cross-linking

and hardening o f the exoskeleton during molting (Terwilliger 1999).

A number o f studies have examined the fate o f Cu and Zn associated w ith the

molt cycle, but report widely different results. Fowleret al. (1971) found that Zn

accumulated from the dissolved phase associated with interstitial spaces in the

exoskeleton and that up to 41% o f the Zn body burden was lost w ith the molted exuviae.

However, several investigators report much higher rates o f retention o f metals during

ecdysis. Fowleret al. (1970) found that Zn accumulated from food became localized on

ligand-binding sites on the underside o f the exoskeleton, and in long-term experiments,

only 1 % accumulated from food was eliminated with each molt. Weeks et al. (1992)

found negligible quantities o f Cu and Zn in the exuviae o f talitrid amphipods. Similarly,

negligible levels o f Cu were found in the exuviae oCallinectes f sapidus (Engle 1987).

Variable results may be due to differences in physiology between species and/or external

metal concentrations.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Metals may also associate with the exoskeleton through adsorptive processes.

Chitin and associated proteins have many hydroxyl, nitrogen, and sulfur-containing

groups that may serve as ligand-binding sites for cationic metals. These sites may be

quite common on the crustacean epicuticle. enhancing the potential for adsorption.

Further, i f molting occurs in contaminated water, metals may adsorb from the aqueous

phase directly to the newly exposed surfaces o f the procuticle o f the exuviae (White and

Rainbow 1984). The adsorbed metals may be bioaccumulated if the organism ingests its

exuviae. Such adsorption o f metals onto binding sites not associated with the epicuticle

o f exuviae may overestimate the role o f ecdysis in the elimination o f metals.

The purpose of this study was to quantify the sequestration o f selected metals (Zn.

Cu. Cd) in sublethal concentrations from the aqueous phase by the exoskeleton o f the

grass shrimp. Palaemonetes pugio Holthius and to determine what fraction o f these trace

metals are depurated through ecdysis. Also because o f its potential importance in

aquaculture (where a high percentage o f crustacean exuviae are ingested), the adsorption

o f metals from the aqueous phase by the exuviae was measured.

METHODS

Exposure of Post-ccdysis Shrimp

In order to determine metal sequestration by the exoskeleton and depuration

through ecdysis. grass shrimp were collected from aSpartina alterniflora salt marsh

(salinity at the time o f collection was 15 %o) near Cocodrie. La. Grass shrimp from 22-28

mm in length from rostrum to telson were retained to maintain a consistent surface-area-

to-volume relationship and molt frequency among individuals. Shrimp were acclimated

to a salinity o f 18 o/00 jn the laboratory for 24 h prior to the experiment.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Single shrimp were placed in 50 low-density polyethylene (LDPE) acid-cleaned

chambers in 18 %o Instant Ocean0 (World Aquarium Systems) artificial seawater (ASW).

exposed to aqueous sublethal mixtures o f elemental Cd. Cu, and Zn (200 pg L '1 Cd, 1000

pg L-1 Zn. 500 pg L’1 Cu). and examined four times daily for molting. The sublethal

mixture o f the three metal concentrations was chosen based on unpublished data (Keteles,

personal observation). Exuviae from shrimp that molted in metal-contaminated water

were removed and analyzed for metal content. After 72 h, unmolted shrimp were

removed and placed in LDPE acid-cleaned chambers containing clean 18 %o ASW

without elevated metals; shrimp molted an average o f 3.8 (± 0.35 SEM) days after being

placed in uncontaminated water. After ecdysis in uncontaminated water, the exuviae and

corresponding whole shrimp were frozen in liquid nitrogen and later analyzed for metal

content. Metal concentrations were measured in six replicates o f three randomly pooled

whole shrimp and corresponding exuviae.

An estimate o f percent depurated from a single molt was calculated based on the

mass o f metals retained in whole shrimp after molting compared to the mass o f metals

eliminated with the exuviae o f the post-ecdysis shrimp. In order to determine the

magnitude o f the adsorption o f metals onto exuviae, metal concentrations in exuviae cast

in contaminated water were compared with exuviae cast in uncontaminated water and to

concentrations in intermolt exoskeletons.

Exposure of Intermolt Shrimp

Metal concentrations were also measured in metal-exposed shrimp that either did

not molt or were continuously exposed before and after a molt to determine the

importance o f the exoskeleton in the sequestration o f metals and to compare to post-molt

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. metal-exposed shrimp. Twenty grass shrimp were exposed to the same sublethal mixture

o f metals (200 pg L '1 Cd, 1000 pg L’1 Zn. 500 pg L '1 Cu) in each o f 6 replicate 4-L jars in

18 %o ASW and monitored daily for indications o f molting. A small fraction (10%) of

the grass shrimp in each jar molted. The low percentage o f molts suggested that metal

uptake in these shrimp would represent metal burdens under intermolt conditions.

Following 72 h o f exposure, shrimp were frozen in liquid nitrogen and exoskeletons

(excluding legs) were excised. Metal concentrations were measured in the excised-

exoskeletons and the remaining tissues o f 10 pooled shrimp from each jar.

Intermolt vs. Post ecdysis Comparisons

Data from the post-ecdysis and intermolt experiments were used to calculate

composite whole-body metal burdens in order to determine i f body burdens were reduced

after a single molt. Excised-exoskeleton metal concentrations o f intermolt shrimp were

compared with metal concentrations in the exuviae from post-ecdysis shrimp which

molted in uncontaminated water.

Metal Analysis

Samples were placed in preweighed, acid-cleaned glass test tubes, weighed, and

digested in 5 ml o f concentrated trace metal grade HNO, (Fisher Scientific) at 90 C for

12 H. The acid containing digested samples were evaporated to about 1.5 ml. then diluted

to -25 ml and reweighed to determine volume. Blanks were prepared for each set o f

samples digested. Metal concentrations in each fraction were determined by Inductively

Coupled Plasma (ICP) emission spectroscopy and corrected for interelement interference.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Statistical Methods

Whole-body metal burdens o f post-ecdysis grass shrimp were compared with

intermolt whole-body burdens using a t-test to determine if concentrations o f Cu, Cd, and

Zn were altered after a single molt. The concentrations o f Cu, Zn. and Cd in the post-

ecdysis grass shrimp exuviae produced in both contaminated and uncontaminated water

were also compared with concentrations in the excised exoskeletons o f intermolt grass

shrimp using a t-test. Exuviae produced in uncontaminated water were compared with

exuviae from contaminated water. In some instances, comparisons failed normality; if so

the nonparametric Mann-Whitney Rank Sum Test was used. Sample size for each metal

was 6 replicates, and all tests were conducted with Sigma Stat (Jandel Scientific*)

software.

RESULTS

Whole-body Burdens of Intermolt and Post-ecdysis Shrimp

Trace meial whole-body burdens o f intermolt P. pugio exposed to elevated metals

averaged 94. 2.9. and 158 |ig g'1 for Cu. Cd. and Zn respectively (Figure 1). Metal

concentrations in intermolt shrimp were compared to whole-body burdens in post-ecdysis

shrimp (exposed to elevated metals for three days and allowed to molt in “clean” water) to

estimate the retention o f metals through a molt cycle. Results varied among the metals

tested. Cu whole-body burdens in post-ecdysis shrimp averaged 80 pg g'1 (Figure 1). Cu

concentration did not decrease significantly after a single molt (p=0.209; power=l .000);

11% of the Cu intermolt body burden was associated with the exuviae (Figure 2). Cd

whole-body burdens after ecdysis (Figure 1) averaged 1.7 pg g'1 and Cd whole-body

burdens decreased significantly after a single molt (p=0.049); 26% o f the intermolt Cd

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 125 i

100

75 H

175

i i Intermolt V////A post-ecdysis

Figure I. Mean whole-body burdens (p g g ')o f Cu, Zn. and Cd o f intermolt and post-ecdysis Palaemonetes pugio exposed to metal-enriched water. Post-ecdysis shrimp molted in uncontaminated water (n=6). Intermolt values were based on a composite shrimp, combining the total metals in the exoskeleton and tissues (n=6). Error bars represent standard error o f the meaa

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 40 -

«4—» C

0 20 -

Cu Zn Cd

F ig u re 2. Depuration as a function o f the relative percent o f the whole-body burden that was eliminated with the exuviae in the grass shrimp, Palaemonetespugio. Percent was calculated based upon the total mass (pg) o f metal associated with the whole-body burden and the exuviae.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. body burden was eliminated with the exuviae (Figure 2). Whole-body burdens o f Zn in

post-ecdysis shrimp averaged 43 pg g'1 (Figure I ), which was significantly lower than

intermolt concentrations (p= <0.001) even though only 18% o f the intermolt body burden

was associated with the exuviae (Figure 2).

Concentrations in Exoskeleton and Exuviae

IntermoltP. pugio stored a substantial portion o f their total metal body burden in

the exoskeleton. The exposure to elevated aqueous metals (500 pg L '1 Cu. 1000 pg L 1

Zn. and 200 pg L 1) resulted in exoskeleton concentrations o f 91 pg g-1 Cu. 2.9 pg g'1 Cd.

and 220 pg g'1 Zn (Figure 3) which represented 36. 52 and 40 % respectively o f the total

Cu. Zn. and Cd body burden (Figure 3). Metal concentrations o f the exuviae produced

after exposure to elevated metals but allowed to molt in clean water averaged 15. 0.8 .

and 15 pg g'1 Cu. Cd. and Zn respectively and were compared to the concentrations in

excised exoskeletons o f intermolt shrimp (Figure 3). With the exception o f Cd, the

intermolt exoskeleton concentrations were significantly higher than the concentrations in

the exuviae (p= 0.001 for Cu. 0.061 for Cd (power=0.397). and <0.001 forZn).

Adsorption onto Exuviae

Metal concentrations o f exuviae produced in contaminated water averaged 198

pg g'1 Cu. 36 pg g 1 Cd. and 382 pg g'1 Zn (Figure 3). Each was significantly higher than

in exuviae produced by molting in uncontaminated water (p= 0.001 for Cu. 0.037 for Cd.

and 0.007 for Zn), suggesting that metals adsorb to the exposed surfaces o f the exuviae.

When exuviae produced by molting in contaminated water were compared to excised

intermolt exoskeletons, only exuvial Cd concentrations were significantly higher

(p=0.018). Cu and Zn concentrations o f exuvia cast in contaminated water were elevated

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. i = Intermolt Exoskeleton I eszzzi Exuviae in Uncontaminated Water I -'^-52 Exuviae in Contaminated Water

Figure 3. MeanCu. Zn. and Cd concentrations o f Palaemonetespugio intermolt exoskeletons from metal-exposed unmolted shrimp (n=6), and post-ecdysis exuviae cast in metal- contaminated (n=6) and uncontaminated (n=6) water. Error bars represent the standard error o f the meaa

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. compared to excised intermolt exoskeletons, but were not significant at the 0.05 level

(p=0.059 for Cu and p=0.087 for Zn).

DISCUSSION

Our results were designed to determine the role o f ecdysis in the fate o f metals in

Palaemonetes pugio. Elimination o f metals in decapod crustaceans can be attributed to

excretion via the urine and feces, flux across the body surfaces and gills (Bryan 1968). as

well as with exuviae. Because the grass shrimp in our post-ecdysis experiment were held

in seawater without elevated metals for an average o f 3.8 d prior to molting, some metals

taken up during the exposure period were likely excreted. If excretion was important to

the fate o f metals, our estimates o f loss through the exuviae w ill be underestimated.

Therefore, data on loss to exuviae were used in concert with whole-body burden

comparisons in intermolt and post-ecdysis grass shrimp to determine if there was a

potential for loss by a mechanism other than molting. Our results suggest that the

potential role for depuration with the exuviae differs for Cu. Cd. and Zn.

Cu whole-body burdens o f intermolt and post-ecdysis shrimp were not

significantly different, and Cu concentration in exuviae was significantly lower than in

the intermolt exoskeleton (Figures 1 and 3). These data suggest that grass shrimp

conserve Cu by reabsorption from the exoskeleton prior to molting, and that relatively

small amounts o f Cu are excreted. Negligible quantities o f Cu were associated with the

exuviae in talitrid amphipods (Weeks et al. 1992) and the blue crab. Callinectes sapidus,

(Engle 1987) further suggesting a conservation o f Cu during the molt cycle. Research

with the blue crab. Callinectes sapidus (Engle 1987), and the lobster. Homarns

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. gammams (Hagerman 1983), showed that the distribution o f Cu among tissues is altered

just prior to ecdysis. Post-ecdysis C sapidus may use stored Cu to synthesize

hemocyanin (Engle 1987) and P. pugio may similarly utilize Cu conserved during ecdysis

in hemocyanin production. Furthermore, the crustacean exoskeleton is formed by

hypodermis cells, and hemolymph proteins similar to hemocyanin are instrumental in the

synthesis o f the new exoskeleton (Terwilliger 1999). Cu containing prophenoloxidases

cross-link and harden the exoskeleton during ecdysis.

Our results for Cd suggest that intermolt exoskeletonsi o P. pugio sequester a

relatively large fraction (40%) o f the total Cd whole-body burden. Similar results were

found by Khan el al. (1989) in P.pugio collected from a contaminated site. The total Cd

whole-body burdens in P. pugio were significantly lower in post-ecdysis than in intennolt

shrimp in our study (Figure 1). Additionally, there was no difference between exuviae

and intermolt exoskeleton Cd concentrations (Figure 3). and 26 % o f the intermolt Cd

body burden was associated with the exuviae (Figure 2). Taken together, these data

suggest that loss to the exuviae substantially contributed to Cd depuration, but do not

exclude some loss by other mechanisms. A 50% loss o f Cd by excretion was measured in

pink shrimp. Penaeus duaromum. in 7 to 10 days (Nimmo et at. 1977). Furthermore, it is

possible that some Cd was mobilized and transported from the exoskeleton to tissues

prior to ecdysis. Nimmoet at. (1977) found that when transfered to Cd-free water

following exposure. Cd levels in pink shrimp exoskeieton decreased, while the

concentration in muscle increased. Sensitivity to metal toxicity has been attributed to the

uptake o f ions from the external medium during the post-molt period in the shore crab.

Carcinus maenas (Scott-Fordsmand and Depledge 1997): however, the results o f our

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. study suggest that high mortality o f molting crustaceans during toxicity bioassays could

be attributed to an increase in metal levels by redistribution to tissues preceding ecdysis.

Zn whole-body burdens were significantly lower in post-ecdysis P. pugio than in

the intermolt grass shrimp (Figure 1). Moreover, exuvial Zn concentrations were

significantly lower than that o f intennolt exoskeletons (Figure 3) and 18% o f the Zn

whole-body burden was associated with the exuviae (Figure 2). Therefore, it appears that

Zn was depurated, but our data suggest that excretion is probably more important to the

fate o f Zn than the loss with exuviae. Fowleret cil. (1971) reported that molting

accounted only for a 1% loss o f ^'Zn from the initial body burden over five months for a

euphausiid. yet 96% o f the initial body burden was eliminated, suggesting excretion was

the primary route o f loss. Large losses o f Zn in decapod crustaceans have been attributed

to excretion across the gills (Bryan 1968). Decapod crustaceans are typically effective

regulators o f Zn content; however. Zn flux can vary w ith external Zn concentrations.

temperature, and between individuals (White and Rainbow 1984). Alkaline phosphatase.

a Zn metal loenzyme. has been found to be associated w ith the new exoskeleton o f a

freshwater crayfish before ecdysis (Bryan 1968). and it is possible that molt physiology

may influence Zn flux in crustaceans. * Significant adsorption to exuviae occurred from aqueous metals as suggested by

the elevated levels o f Cu. Cd. and Zn associated with molted exuviae in metal-enriched

water (Figure 3). Cd concentrations o f exuviae cast in contaminated water were also

significantly higher than intermolt exoskeletons. Chitin and associated proteins have

many hydroxyl, nitrogen, and sulfur-containing functional groups that may serve as

ligand-binding sites for cationic metals (Stevenson 1985). Many such sites are likely to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. be quite common on the crustacean epicuticle, enhancing the potential for adsorption.

These ligand-binding sites may be even more common on the underside o f the molted

exoskeleton. Therefore, after molting, new surfaces for adsorption could contribute to the

increase in the concentrations o f exuviae cast in contaminated water. White and Rainbow

(1984) also found that crustacean exuviae take up Zn. Cu. and Cd from the ambient

medium. The adsorption o f metals onto the shed molt maybe an important phenomenon

in the cycling o f trace metals in aquatic systems (Wang et al. 1996). Additionally, many

crustaceans, particularly in densely-populated farming conditions, ingest exuviae in order

to recycle nitrogen and compensate for the energetic loss o f exuvial production

(Guillaume 1997). In fact, the molted exoskeletons are incorporated into the calculated

food budgets o f commercial shrimp production (Dr. R. Portier. personal communication).

Elevated levels o f metals associated with molted exoskeleton due to adsorptive processes

under contaminated conditions, as a result o f the ingestion o f exuviae, may be an

important source o f metal bioaccumulation by crustaceans.

Our work suggests that a substantial portion o f the whole-body burdens o f Cu. Zn

and Cd are associated with the exoskeleton (36, 52 and 40%. respectively) and that the

molt cycle is important to the fate and distribution o f metals in P. pugio. Ecdysis

contributes to reductions in Cd burdens while Zn appears to be largely depurated by

excretion. Most Cu associated with the exoskeleton appears to be redistributed to the

tissues before molting and is therefore retained after a molt. Furthermore, metals appear

to adsorb to the surface o f P. pugio exuviae when cast in metal-enriched water. Such

findings may have important implications for crustacean-metal interactions. First, metals

may not exert a toxic effect when associated with the exoskeleton. This suggests that

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. critical-bocfy residue p*edictions of toxicity derived in other t&xa may be inaccurate in

crustaceans and that crustaceans may not serve as reliable biomonitors. Our data also

suggest that tissue levels vary over a molt cycle, possibly influencing the timing o f toxic

effects. Second, metal bioavailability in tissues and exoskeleton may differ, influencing

trophic transfer o f metals to predators.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER3

METAL UPTAKE AND PARTITIONING AMONG TISSUES AND EXOSKELETON OF THE GRASS SHRIMP, PALAEMONETES PUGIO

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. INTRODUCTION

Trace metals have been shown to associate with both tissues and the chitinous

exoskeleton in crustaceans. Measurements o f a variety o f metals(e.g., Cu. Zn. Cd. Co,

Se. Hg) associated with exoskeletons have been made in at least eight species o f

crustaceans in groups as diverse as decapods and copepods (Zanders and Rojas 1996;

Khan et cil. 1989; Mungerand Hare 1997; Depledge et al. 1993; Reinfelder and Fisher

1994; White and Rainbow 1984; Nimmo et al. 1977: Fowler et al. 1970; Bryan 1968).

Several studies (Zanders and Rojas 1996: Khan et al. 1989; Reinfelder and Fisher 1994:

Fowleret al. 1970: Depledge et al. 1993; Nimmo et al. 1977) have found higher

concentrations of Cd and Zn in the exoskeleton compared to whole-body tissues (or

specific tissues such as muscle). The highest reported values (97% o f the total body Cd.

Zn. and Co burden) o f the exoskeleton have come from marine planktonic copepods

(Reinfelder and Fisher 1994): however, a freshwater cladoceran was shown to have 19%

o f its Cd burden in the exoskeleton (Munger and Hare 1997). O f the metals studied. Cu

appears to have the lowest fraction associated with the exoskeleton (Depledgeet al.

1993). Variability among these studies could be due to differences in physiology, metal-

species behavior, salinity, and source o f exposure.

Metals may associate w'ith the exoskeleton in two ways. First, the deposition o f

calcium salts following molting may facilitate the sequestration o f trace metals within the

procuticle (or matrix) o f the exoskeleton (Adams 1981. Stevenson 1985). White and

Rainbow (1984) found that zinc concentration in the cuticle varied with degree of

calcification in a fashion closely related to the molt cycle. Cations bind to chitin by

forming ionic bonds with nitrogen side groups within the chitin fibrils o f the procuticle.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Some fraction o f these cationic metals, however, may be subject to mobilization and

release to the tissues along with Ca prior to molting. Other metals become associated

with the exoskeleton through different mechanisms, but may also be subject to

physiological variation associated with the molt cycle. For example, copper-containing

proteins similar to the oxygen-carrier pigment hemocyanin are involved in cross-linking

and hardening o f the exoskeleton during molting (Terwilliger 1999).

The second method o f metal association with the exoskeleton is through an

adsorptive process. Chitin and associated proteins have many hydroxyl, nitrogen, and

sulfur-containing groups that could serve as ligand-binding sites for cationic metals

(Stevenson 1985). Such sites are likely to be quite common on the crustacean epicuticle.

enhancing the potential for surface adsorption. It appears that much of the zinc associated

with the exoskeleton o f some decapod crustaceans is adsorbed to the surface, and metal

concentration o f the exoskeleton increases when the concentration o f the ambient water

increases (Bryan 1968). Adsorption to exuviae is know to occur from metal-enriched

water (Keteles and Fleeger. in press). If significant adsorption on the epicuticle occurs,

surface area may influence the quantity o f adsorbed contaminant per individual

crustacean, and body size may therefore contribute to species-specific differences in

adsorption. Exoskeleton chemical composition also differs among crustaceans, adding to

the potential for species-specific differences in contaminant partitioning. Furthermore,

adsorption o f contaminants likely occurs directly to the epicuticle without association

with biological tissues, and such adsorbed substances may not contribute to toxic effects.

Although many authors (Reinfelder and Fisher 1994; Zanders and Rojas 1996;

Smokerowskiet al. 1998) acknowledge the potential significance o f adsorption, no

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. studies have attempted to directly quantify the distribution between metals adsorbed to

the surface and metals sequestered within the crustacean chitin matrix in the procuticle,

although studies have show that 7-40% is adsorbed to the exoskeleton o f aquatic nymphs

(Hare 1992).

Partitioning among these various exoskeleton and tissue pools in crustaceans may

vary as a function o f exposure route (Fowleret al. 1970). Contaminants that enter with

the ingestion o f food, and are subsequently absorbed by the gut. may be subject to

physiological regulating mechanisms (Rainbow 1995). The fate o f such contaminants is

to be stored in various tissue and exoskeleton pools or ultimately metabolized or excreted

(Fowleret al. 1970). Generally, tissue concentrations o f Zn and Cu appear to be tightly

regulated in crustaceans (Bryan 1968; Rainbow' 1995) at least at low exposure

concentrations; Cd does not appear to be regulated (Rainbow 1995). Metals obtained

from food w ill likely associate with the chitin matrix o f the exoskeleton only after

absorption and transport through the circulatory system. Alternatively, some fraction o f

dissolved metals may rapidly reach equilibrium directly by adsorption to the surface o f

exoskeletons without first entering tissues and being subject to regulatory mechanisms.

The objective o f this study was to determine how the source o f exposure (aqueous

or dietary) effects uptake and partitioning o f Cu. Zn. and Cd between the tissues, chitin

matrix in the procuticle, and the surface o f the epicuticle in Palaemonetes pugio Holthius.

Epiphytic algae associated with the stems o f Spar tina allerniflora were used as the food

source. Grass shrimp are known to ingest epiphytes (Morgan 1980) including those

associated with S. allerniflora (Fleegeret al. 1999). Because salinity influences metal

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. speciation. salinity effects on uptake and partitioning from aqueous exposure were also

examined.

METHODS

Aqueous Exposure with Varying Salinity

In order to determine metal bioconcentration and partitioning between

exoskeleton and soft tissues, grass shrimp were collected fromSpartina a allerniflora salt

marsh (at 10 %o and 27°C) near Cocodrie, La on May 21. 1999. An additional collection

was made on August 26. 1999 at the same site (at 15 %o and 29.4° C). Grass shrimp were

subsequently exposed to an aqueous sublethal mixture o f Cd. Cu. Zn (200 pg L '1 Cd.

1 0 0 0 pg L’1 Zn. 5 0 0 pg L 1 Cu) in artificial seawater (ASW) comprised o f ultra clean

water (M illi-Q ) and Instant Ocean sea salts for three days. This sublethal mixture o f the

three metal concentrations was used in previous experiments and was found to lead to

metal uptake without elevates mortality (Keteles and Fleeger. in press). Experiments

were conducted at 5. 18 and 3 0 % o. and shrimp were step-wise acclimated at a rate

change of 5 %o per day to the desired salinity. Twenty grass shrimp per replicate were

exposed to the metal mixture in3 L o f ASW in acid-cleaned 4-L glass jars. Four

replicate jars w ere used at 5 and 30% o because o f the limited number o f shrimp available

(collected on August 26. 1999); six replicates were used at 18% o (collected on May 2 1 .

1999). Jars were covered with plastic wrap to prevent evaporation, and each was aerated

by airstone to maintain dissolved oxygen at saturation. Twenty shrimp were also placed

in one jar with ASW (no metals amended) o f the appropriate salinity for each experiment

to determine background burdens. Shrimp were selected on the basis o f size (all were 22-

28 mm in length from rostrum to telson) to maintain a consistent surface-area-to-volume

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. relationship and molt frequency, and shrimp were monitored daily for molting during the

course o f the experiment. Approximately, 10% o f the grass shrimp molted during the

course o f the experiment.

Dietary Exposure

Grass shrimp were later collected at the same sampling site on May 6. 2000 (at

15%o and 2 3 ° C) and acclimated to 18 %o. P. pugio (again ranging from 22-28 mm) were

fed Zn. Cu. and Cd-enriched epiphytic microalgae to determine the uptake and

partitioning o f metals from dietary exposure. Epiphytes associated with S. alterniflora

stems were exposed to Cd. Cu. and Zn concentrations o f 200, 500 and 1000 |ig L '1

respectively amended in 18%o clean ASW and placed under cool white lights ( 1 4 :8 light:

dark cycle). Epiphyte exposure was conducted in staggered batches o f one-day intervals

to maintain a consistent exposure o f epiphytes to be fed to shrimp. Following three days

o f exposure, stems were removed, rinsed with ethylenediaminetetraacetate (EDTA) to

remove metals adsorbed to the surfaces o f algal cells and stems. A subsample o f

epiphytes scraped from three stems with a razor blade was frozen and retained for

analysis o f metals to determine uptake by the microalgae.

Five groups o f 20 shrimp were placed in3 - L acid-cleaned glass fish bowls and

allowed to feed daily on either metal-enriched or uncontaminated epiphytic microalgae

(control) for three days in 2 L o f 1896o clean ASW. Five epiphyte laden stems

(approximately 10 cm in length) were held vertically by plastic mesh placed at the bottom

o f each fish bowl. Shrimp were allowed to feedad libitum on epiphytes each day for 12 h

after which stems (and remaining microalgae) were removed. Water was replaced daily

3 4

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. to lim it the aqueous exposure from metals depurated from the shrimp or microalgae

during the course o f the experiment.

Post-exposure Methods

Grass shrimp from each experiment described above were removed following

exposure and sacrificed by freezing in liquid nitrogen. H alf (ten) o f the grass shrimp from

each jar were rinsed three times for 20 min with 20 ml o f 10 mM EDTA to chelate and

remove metals adsorbed to the surface o f the epicuticle. Exoskeletons were peeled from

remaining tissues (excluding legs and antennae) o f the EDTA-rinsed (referred to hereafter

as exoskeleton matrix) and the unrinsed (referred to hereafter as total exoskeleton)

shrimp. Cu. Zn. and Cd were measured in the exoskeleton and soft tissues o f both the

EDTA-rinsed and the unrinsed shrimp in aggregate, i.e. measurements were made on 10

pooled exoskeletons or tissue from a single jar. The dry mass o f exoskeletons and tissues

from each aggregate was weighed with a Mettler balance (model pm460. ± 1 pg).

Metal concentrations (in pg g'1) were measured in three shrimp compartments:

total exoskeleton, exoskeleton matrix and tissue. Whole-body burdens were estimated by

summing tissue and total exoskeleton values. For tissues and exoskeleton, the mass o f

each metal (per shrimp) was estimated from the concentration and appropriate body mass.

Metals adsorbed onto the epicuticle surface were determined by calculating the difference

in mass (per shrimp) o f each metal in the unrinsed exoskeletons and exoskeletons rinsed

with EDTA. Partitioning o f metals was also measured in control grass shrimp (not

exposed to elevated metals levels) in order estimate background burdens.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Enrichment Factors

Cu, Zn, and Cd concentrations in shrimp not exposed to elevated metals (control

shrimp) were subtracted from the concentrations in the tissue and exoskeleton

compartments o f metal-exposed shrimp to calculate metal bioconcentration and

bioaccumulation factors. The degree o f concentration o f Cu. Zn. and Cd in the shrimp

was calculated as an enrichment factor for the tissues and exoskeleton at each salinity

and for both sources o f exposure. Enrichment factors o f dietary (bioaccumulation factor)

and water-only (bioconcentration factor) exposure for tissue, total exoskeleton, and

exoskeleton matrix were determined and compared. Bioconcentration factors o f the three

compartments were also compared for each salinity.

Enrichment factors are defined as:

Bioconcentration Factor

metal concentration in biogenic material (corrected for background) metal concentration in seawater

Bioaccumulation Factor

metal concentration in biogenic material (corrected for background) metal concentration in food.

Metal Analysis

Shrimp tissues or exoskeletons were placed in preweighed, acid-cleaned glass test

tubes, weighed, and digested in 5 ml o f concentrated trace-metal-grade H N 03 (Fisher

Scientific) at 90° C for 12 h. The acid-containing digested samples were evaporated to

about 1.5 ml. then diluted to -25 ml and reweighed to determine volume. Blanks were

prepared for each set o f samples digested. Cu, Zn, and Cd concentrations in each fraction

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. were determined by Inductively Coupled Plasma (ICP) emission spectroscopy and

corrected for interelement interference.

Statistical Methods

The effects o f source (dietary and aqueous exposure) and salinity on metal

partitioning were addressed in two ways. First, metal burden in tissues and total

exoskeleton was analyzed by one-way ANOVA. Due to lack o f independence, the

difference between tissues and total exoskeleton (pg shrimp ') from each replicate was

calculated. This difference was compared at each salinity to determine i f salinity

influenced the fate o f metals. In a separate one-way ANOVA. source (dietary vs. 18%o

aqueous uptake) effects on the difference between tissue and total exoskeleton burdens

were also compared. Second, two-way ANOVA's were conducted to determine i f

salinity or source altered partitioning o f metals associated only with the exoskeleton.

Metal burdens (pg shrim p1) o f the exoskeleton o f EDTA-rinsed and unrinsed shrimp

were treated as a main effect. Salinity and source (dietary vs. 18 %o aqueous uptake) were

the other main effect in separate two-way ANOVA's. Interactions were also examined.

Significant differences (p<0.05) were analyzed with Tukey's post-ANOVA analysis.

RESULTS

Metal Concentrations

Control shrimp used in both experiments contained relatively high background

whole-body burdens o fZ n and Cu. Background Cd was, however, differed substantially

between aqueous and dietary-exposure experiments, perhaps because shrimp were

collected at different times. However, whole-body burdens after exposure to elevated

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. metals from food or water were always higher than in shrimp not exposed to elevated

metals (Figure 4).

When exposed to elevated aqueous metals for three days, grass shrimp Cu. Zn.

and Cd whole- body concentrations averaged 98. 159. and 3.9 pg g'1 respectively across

all three salinities (Figure 4). Generally, tissue and total exoskeleton concentrations were

broadly overlapping for each metal (data not reported) and served as the basis for

estimating enrichment factors and mass calculations used to calculate metal partitioning.

S. allerniflora epiphytes accumulated metals, and metal-enriched epiphyte

concentrations o f Cu. Zn and Cd were 25 . 38 and 11 pg g'1 respectively (Figure 5).

Whole-body burdens in grass shrimp exposed to metals from ingestion o f epiphytic

microalgae for three days averaged 74. 99. and 3.4pg g-1 for Cu. Zn, and Cd, respectively

(Figure 4).

Bioconcentration and Bioaccumulation

Enrichment factors were calculated for tissue, total exoskeleton and exoskeleton

matrix compartments from aqueous exposure at three salinities and from dietary

exposure. Bioconcentration factors (from aqueous exposure) ranged from4 .4 - 1 0 2 .8 for

Cu. 0 - 1 6 0 .7 for Zn. and 8 .6 -3 1 .1 for Cd (Figure 6 ) . Bioconcentration factors for Zn

tended to be higher at 3 0 %o than at 5 or 18 %o; Cu bioconcentration factors tended to be

lower at 3 0 % o than at 5 or 18 96o. Tissue and total exoskeleton bioconcentration factors

were variable among metal species and condition. Exoskeleton Zn bioconcentration was

higher than tissue bioconcentration at each salinity; however. Cd and Cu bioconcentration

o f the tissue and total exoskeleton varied with salinity. Rinsing exoskeletons w ith EDTA

decreased bioconcentration factors by 0 - 5 0 % . At 5% o. rinsed Cu exoskeleton

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 125

!1— □ Control ! V777Z& Metal-Exposed

8 Cd

0 5 18 30 Food

F ig u re-4. Cu (a). Zn (b). and Cd (c) vvhole-bodyburdens (pgg ‘ ) in P./Ti/g/o vvhole-body (tissue andtotalexoskeleton)arerepresentedincontrolandmetal-exposedgrassshrimpfronr three salinities(aqueousexposure)andexposurefronvfbod.“*

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. i ~=i control metal-exposed

40 O) 30 U i 20

0 Cu Zn Cd

Figure 5. Epiphyte metal concentrations (pg g'1 ) fromSpartina alterniflora stems either exposed or not exposed (control) to elevated concentrations ofCu. Zn. and Cd. N=3, error bars represent standard error o f themeaa

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 200

175

o Cu O « 150 (0 u. U. e e o 125 o

100 o> E o 3 c 75 O o u o n o 50 o m ffi 25

0 .CD 5 ppt 18 ppt 30 ppt

200 20

175 T issu e Zn 16 3 o 150 u u.ra UL<0 Exo e 125 c o 12 o E xo M atrix IS is 100 n s 3"' T is s u e R 5 o 75 o E x o R e o u 50 o o E xo M atrix 25 1 0 5 ppt 18 ppt 30 ppt Food

50 1 00

0 90

2 40 Cd 0 80 o u n n 0.70 u . c c o 30 0 60 o is 2 0 50 « s 3 § E o 20 0 40 3 e oU o 0.30 n 8 o s 10 irr 0 20 0 10

0 00 5 ppt 18 ppt 30 ppt Food

Figure 6.Cu (a), Zn (b), and Cd (c) enrichment in P. pugio. Exoskeleton - adsorbed, total exoskeleton, and tissue values are represented across three salinities (bioconcentration) and from dietary exposure (bioaccumulation, right Y axis (R)). (N=4, 5, 30, and Food; N=6 18) Error bars represent standard error o f the mean.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. bioconcentration was higher than unrinsed, probably because o f background variability

among different collections o f shrimp. Therefore, Cu adsorption to exoskeleton was

considered negligible at 5%o.

Bioaccumulation factors (from dietary exposure) averaged 1.39 for Cu, 0.29 for

Cd, and 4.70 for Zn (Figure 6). Cu and Zn bioaccumulation factors were higher in

tissues than in exoskeleton; bioaccumulation o f Cd however was higher in the

exoskeleton. The EDTA rinse decreased Zn and Cu bioaccumulation by about 90%,

suggesting that a significant fraction o f exoskeleton Cu uptake was by adsorption. EDTA

rinse decreased Cd bioaccumulation by about 30%.

Body Mass

Over both experiments, the exoskeletonP. opugio f averaged 37 % o f the total dry

weight o f grass shrimp. The mean mass o f the shrimp used in these experiments ranged

from 30-50 mg dry wt shrimp'1 and a one-way ANOVA suggested that mass did not vary

among the four salinity or source treatments (p=0.127). Therefore, we felt justified in

comparing the metal mass (on a per shrimp basis) among experiments because we used a

limited size range o f grass shrimp, and because no differences in body mass were

detected among exposure experiments.

Whole-bodv Comparisons

Across all salinity and source treatments, the total mass o f Cu averaged about 3.2

|ig shrimp'1 (Table 1). A range o f62-68 % o f the whole-body Cu burden was associated

with the tissues across the three salinity treatments from aqueous exposure, and 65% o f

the total body burden was associated with the tissues from dietary exposure. Seventeen to

thirty-two percent o f the whole-body Cu burden was associated with the exoskeleton

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 1. The whole-body mass and percentage o f the whole-body burden in P. pugio associated with the tissue, exoskeleton matrix, and adsorbed fractions o f Cu, Zn. and Cd for each salinity treatment and dietary exposure.

Cu pg shrimp'1 Tissue (%) Exoskeleton Exoskeleton Matrix (%) Adsorbed (%)

5% o 3 .9 1 6 8 3 2 0

18% o 4 .4 9 6 4 19 17

3 0 % o 1 .3 0 6 2 17 2 0

Diet 3 .2 7 6 5 17 16

Zn pg shrimp'1 Tissue (%) Exoskeleton Exoskeleton Matrix (%) Adsorbed (%)

5% o 4 .9 2 6 5 3 2

18% o 7 .5 7 4 8 4 5 6

3 0 % o 3 .1 8 5 4 3 9 7

Diet 4 .3 6 5 8 2 4 18

Cd pg shrimp'1 Tissue (%) Exoskeleton Exoskeleton Matrix (%) Adsorbed (%)

5% o 0 .2 9 5 4 4 3 4

18% o 0 .1 4 6 0 2 4 16

3 0 % o 0 .1 0 7 7 7 16

Diet 0 .1 5 3 4 41 2 5

4 3

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. matrix in aqueous exposure, and 17% o f the whole-body Cu burden was associated with

the exoskeleton matrix after exposure from food. Zero to twenty percent o f the whole-

body Cu burden was adsorbed to the exoskeleton after aqueous exposure and 16% was

adsorbed to the surface after exposure from food. No Cu was associated with the

adsorbed fraction at 5%o.

Across all salinity and source treatments, the total mass o f Zn averaged about 5 .0

pg shrimp'1 (Table I ). A range o f 48-65 % o f the whole-body Zn burden was associated

with the tissues across salinity' treatments from aqueous exposure, and 58% o f the total

body burden was associated w ith the tissues from dietary exposure. Thirty-two to forty-

five percent o f the whole-body Zn burden was associated with the exoskeleton matrix in

aqueous exposure, and 18 % o f the whole-body Zn burden was associated with the

exoskeleton matrix after dietary exposure. Three to seven percent o f the whole-body Zn

burden was adsorbed to the exoskeleton after aqueous exposure and 24% was adsorbed to

the surface after dietary exposure. The smallest fraction (3%) was associated with the

adsorbed fraction at 5 %o.

Across all salinity and source treatments, the total mass o f Cd averaged about 0.17

pg shrimp'1 (Table 1). A range of 54-77 % o f the whole-body Cd burden was associated

with the tissues across salinity treatments from aqueous exposure, and 34% o f the total

body burden was associated with the tissues from dietary exposure. Seven to forty-three

percent o f the whole-body Cd burden was associated with the exoskeleton matrix in

aqueous exposure, and 34% o f the whole-body Cd burden was associated with the

exoskeleton matrix after dietary exposure. Four to sixteen percent o f the whole-body Cd

burden was adsorbed to the exoskeleton after aqueous exposure and 25% was adsorbed

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. to the surface after dietary exposure. The lowest percent (4%) was associated with the

adsorbed fraction at 5% o.

Tissue - Exoskeleton Comparisons

Mean tissue Cu ranged from 0.80-2.86 pg shrimp'1, and mean total exoskeleton

Cu ranged from 0.50-1.63 pg shrimp'1 (Figure 7). Salinity did not influence the difference

between tissue and exoskeleton Cu mass (p=0.5390) (Table 2), but source o f exposure did

(p=0.0081, Table 2). The difference between tissue and total exoskeleton Cu in aqueous

exposure (1.30 pg shrimp'1) was greater than in dietary exposure (0.95 pg shrimp'1).

Neither salinity nor source significantly influenced the difference among the tissue and

total exoskeleton compartments o f Zn mass (p=0.2675 for salinity; p=0.2474 for source)

or Cd mass (p=0.1565 for salinity, p=0.0952 for source; Table 2). Mean tissue Zn was 2.8

(± 0.4 standard error o f the mean; this convention w ill be used in all values that follow)

pg shrimp'1 and mean total exoskeleton Zn was 2.2(±0.6) pg shrimp'1 (Figure 7). Tissue

Cd averaged 0.09 (± 0.02) pg shrimp'1 and total exoskeleton Cd averaged 0.08 (±0.02) pg

shrimp'1 (Figure 7).

Intra-Exoskeleton Comparisons

Two-way ANOVA showed that salinity affected the Cu exoskeleton metal mass

such that exoskeleton Cu mass at 5 and 18%o was greater than at 30 %o (p=0.0020. Table

3). Additionally, exoskeletons had significantly more Zn mass at 18 96o than at either 5 or

30 96o (p=0.0006). Salinity did not significantly influence Cd exoskeleton mass

(p=0.2861). The EDTA rinse reduced exoskeleton Cu mass (total exoskeleton Cu burden

was significantly higher than the exoskeleton matrix burden; p=0.0165, Table 3). Total

exoskeleton Cu averaged 1.3 (± 0.2) and exoskeleton matrix values averaged 0.9 (±0.1)

4 5

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o . o

3.0 - Cu

2.5 -

2.0 -

0.5

0.0 5.0 4.5 - Zn 4.0 1 I Q. 3.5 -

2.0 -i

O) 3 . 0.5 -

0.0

0.30 - r

025

0.20 - j

0.15 -

0.10 -i y /. / V/',-

0.05 -i

0.00 - 5 18 30 Food : j Tissue Total Exoskeleton Exoskeleton Matrix Figure 7. Cu (a), Zn (b). and Cd (c) burdens (pg shrimp ') in P.pugio. Tissue, total exoskeletoa and exoskeleton - adsorbed values are from dietary and aqueous exposure (three salinities). N=4 5, 30 and food; n=6 18 and error bars represent standard error o f the mean. 46

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2. One-way ANOVA comparisons o f tissue - total exoskeleton mass for salinity (5.18, and 30 %o) and source (water vs dietary) for Cu, Zn, and Cd.

Metal Source (p) Salinity (p)

Cu Water > Food (0.0081) 0.5390

Zn 0.2474 0.2675

Cd 0.0952 0.1565

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3. Two-way ANOVA comparisons o f Cu, Zn, and Cd exoskeleton mass o f EDTA rinsed and unrinsed exoskeletons across salinities (5, 18, and 30 %o)

Metal (model p) Salinity (p) Rinse (p) Rinse * Salinity (p)

Cu 18>30. 5>30 Unrinsed > Rinsed (0.0233) (0.0020) (0.0165) 0.0919

Zn 18>5 : 18> 30 (0.0345) (0.0006) 0.7020 0.9173

Cd (0.5085) 0.2861 0.9011 0.5084

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. fig shrimp'1 (Figure 7). The EDTA rinse did not reduce the exoskeleton Zn (p-0.7020)

or Cd (0.9011) mass. Interactions between salinity and rinsing were not significant for

Cu (p=0.0919). Zn (p=0.9173) or Cd (p =0.5084).

Exoskeleton Zn and Cu masses were significantly higher when the source of

exposure was from water (p = f .0012 for Zn, p= 0.0044 for Cu: Table 4). Source did not

influence Cd exoskeleton mass (p=0.0694). The EDTA rinse significantly reduced the

exoskeleton Cu mass (p=0.005). but rinsing did not significantly reduce Zn or Cd mass.

Total Cu exoskeleton mass was significantly higher (1.4 pg shrimp'1) than exoskeleton

matrix mass (0.7 pg shrim p1) (Figure 7). Interactions between salinity and rinsing were

not significant for Cu (p=0.4024). Zn (p=0.7754) or Cd (p=853. Table 2).

DISCUSSION

Experiments withPalaemonetes pugio were conducted over a 72 h exposure to

elevated metals, and resulted in uptake o f Cu. Zn. and Cd in tissues and exoskeleton

(surface adsorption and deposition within the chitinous exoskeleton matrix). Most grass

shrimp did not molt during the exposure period, but it is possible that metals, at least in

the tissues and exoskeleton matrix, did not reach equilibrium. Metals adsorbed to the

surface o f the epicutiele in relatively small crustaceans like P. pugio probably equilibrium

quickly (White and Rainbow 1984). Previous work by Nimmo et al. (1977) suggests that

Cd whole-body concentrations reach equilibrium in P. pugio in no more than seven days.

However. Zn and Cu may reach equilibrium sooner because they are tightly regulated in

decapods (Bryan 1968). Because tissue and internal exoskeleton matrix Cd

concentrations may not have reached equilibrium in my experiments, the proportion

adsorbed to the surface may decrease over time if the concentration in the tissues and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4. Two-way AN O VA comparisons o f Cu, Zn. and Cd exoskeleton mass o f EDTA rinsed and unrinsed exoskeletons from dietary and aqueous ( 18 % o) exposure.

Metal (model p) Source (p) Rinse (p) Rinse * Source (p)

Cu Water > Food Unrinsed > Rinsed (0.0142) (0.0044) (0.005) 0.4042

Zn Water > Food (0.1244) (0.0012) 0.2188 0.7754

Cd (0.3366) 0.0694 0.0684 0.8352

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. exoskeleton matrix increase over time. Therefore, i f the tissues did not reach equilibrium

after 96 h. the exoskeleton-adsorbed metals could inflate the proportion associated with

the exoskeleton. However, unpublished data (Keteles) show that field-collectedP. pugio

have a similarly high fraction o f the total metal burden associated with the adsorbed,

suggesting these short-term exposures generate realistic estimates o f exoskeleton values.

Nevertheless, my results appear to be the first to quantify the fraction o f metals adsorbed

to the surface o f the crustacean epicuticle.

The rate and degree at which a dissolved metal adsorbs to organic matter, such as

the crustacean epicuticle, depends upon concentrations o f other solutes (Ca2\ Na '. Cl',

etc.). among other factors (Libes 1992). Salinity greatly influenced adsorbed metal

uptake in P. pugio. The fraction o f adsorbed metal body-burdens was very low at 5 %o

(0-4% o f the total for Cu, Zn, and Cd) perhaps because metals are likely to be in the free

ionic state under conditions o f low salinity. Ninety-seven percent o f the Cd. Se. Co. Zn

body burden was associated with the exoskeleton o f marine zooplankton (Reinfelder and

Fisher 1994). whereas, only 19% of the Cd burden was on the exoskeleton o f freshwater

zooplankton (Munger and Hare 1997). further suggesting that salinity may contribute to

differences in exoskeleton metal content. A t salinities o f 18 and 30%o. the adsorbed

fraction was consistently high and ranged from 16-20% of the whole-body burden for Cu.

6-18% for Zn. and 16-25% for Cd (Table 1). Rinsing with EDTA, a metal chelator,

reduced exoskeleton bioconcentration and biomagnification factors of all three metals by

0-50% (Figure 6). Furthermore, the EDTA rinse significantly reduced the mass o f Cu

associated with the exoskeleton (Table 3 and 4). Although, the EDTA rinse did not

significantly reduce the body burdens o f Zn and Cd (Table 3 and 4). relatively high levels

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o f Zn and Cd were found in the exoskeleton matrix, and thus, the proportion o f adsorbed

Cd and Zn were low relative to the total exoskeleton. Similarly. Hare (1992) reported that

the proportion o f externally bound Cd on an aquatic insect. Hexagenia limbata, decreased

with increasing internal exoskeleton accumulation.

Metals also bioaccumulated and bioconcentrated in the internal exoskeleton

matrix relative to controls, and metals associated with the interior o f the exoskeleton

ranged from 7-45% o f the total metal body burden. Zn and Cd were particularly enriched

in the exoskeleton matrix (Figure 7). A Zn metalloenzyme has been reported within the

exoskeleton o f the freshwater crayfish.Cambarus virilis (Bryan 1968). Because Cd is

chemically similar to Ca. deposition o f Cd in the exoskeleton could occur as grass shrimp

harden the procuticle and epicuticle by the deposition o f calcium salts (Stevenson 1985).

Depledgeet al. (1993) found high levels o f Cd in the exoskeleton o f the benthic crab.

Dorippe granulate, and Khan et al. (1989) similarly found high levels o f Cd in the

exoskeleton oP. f pugio.

Metals externally adsorbed to the surface o f the epicuticle likely have a very'

different potential for impact on the grass shrimp compared with metals integrated into

the exoskeleton matrix. At least a large fraction o f the metals removed by EDTA likely

adsorbed directly from the aqueous solution. Adsorbed metals, therefore, do not have the

opportunity to interact with receptor sites and cause toxic effects. On the other hand,

matrix-bound metals must be transported to the exoskeleton through the circulatory'

system and may interact with tissues (Bryan 1968: Fowler et. al 1970). This suggests that

adsorbed metals inflate estimates o f metal body burden in crustaceans. As a result,

critical body residue predictions o f toxicity for marine crustaceans o f a size similar to P.

Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission. pugio may be underestimated by as much as 15-20% (about the fraction adsorbed in

salinities greater than 5%o). Furthermore, predictions concerning bioavailability o f metals

to crustaceans which do not account for the adsorbed fraction may be overestimated, and

crustaceans may represent poor choices in biomonitoring programs (Langston and Spence

1995). Routine removal o f adsorbed metals with a strong chelator, such as ETDA. before

chemical analysis may provide a more accurate estimate o f total metal burden. Finally,

very little is known concerning the role o f surface-adsorbed metals in trophic transfer o f

metals to higher trophic levels. Because adsorbed metals represent an important part o f

the whole-body burden, surface-adsorbed metals could be an important vector to higher

trophic levels. In fact surface-adsorbed metals may be easily desorbed in the digestive

tract o f a predator.

It has often been suggested that crustaceans store toxic metals in the exoskeleton

and thereby reduce exposure to the organism (Depledgeet al. 1993: Khan et al. 1989:

Reinfelder and Fisher 1994). Flowever, once associated with the exoskeleton matrix,

both essential (Cu. Zn) and nonessential (Cd) metals maybe remobilized during the molt

cycle for storage in tissues for reabsorption following ecdysis. Therefore, tissue metal

concentrations in crustaceans have been shown to vary before and after ecdysis (Depledge

et al. 1993; Nimmo et al. 1977). My previous work suggests large differences among

metals in the potential for depuration in P. pugio through ecdysis: most o f the

exoskeleton Cu was retained in the tissues during a molt, and although a significant

fraction o f the exoskeleton Cd was eliminated with the exuviae, some appears to have

entered tissues (Keteles and Fleeger. in press). This remobilization o f metals from the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. exoskeleton to tissues during molting may cause spikes in tissue concentrations, perhaps

exceeding the critical body residue and causing mortality.

The proportion o f the total Cu and Zn body burden in P. pugio associated with

tissues was relatively high and showed little variability among the salinity and dietary-

exposure treatments. Tissues contained the majority o f the Cu and Zn burden (Table 4).

Zn and Cu are physiologically regulated by crustaceans (Bryan 1968: Rainbow 1995:

Hebei et al. 1997). For example. Cu plays a key role in the oxygen-binding pigment,

hemocyanin. However, exposure to excess Cu activates detoxification mechanisms that

lead to increases in granule formation to aid in excretion (Hebei et al. 1997). Similarly.

Zn is an essential cofactor in enzymatic function, although excess zinc is excreted (Bryan

1968). Therefore, regulation o f Zn and Cu in the tissues may reduce tissue variation

when exposed to elevated Zn and Cu. Cd tissue burden, however, ranged from 34-77 %

o f the whole body burden (Table 4). Cd is a nonessential toxic metal and is not

physiologically regulated (Smokerowski et al. 1998). Therefore. Cd tissue concentrations

may be more variable depending on exposure regime and stage o f the molt cycle.

When metal exposure was through diet, the removal o f adsorbed Zn and Cu

decreased bioaccumulation values by up to 90% and removal decreased Cd by about

30%. Additionally. 16% o f total Cu. 24 % o f total Zn. and 25 % o f total Cd burdens were

adsorbed to the exoskeleton when the source o f exposure w as dietary. These values

generally exceeded surface adsorption associated with aqueous exposure. Several factors

may have contributed to this observation. Metals may have leached from the epiphytic

algae or Spartina stems, been released by sloppy feeding by the shrimp or excreted by the

shrimp into the water, and then adsorbed onto the exoskeleton from the aqueous medium.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Metal concentrations were not measured in the water containing epiphytes and grass

shrimp (although water was replaced daily to minimize aqueous exposure). Wang et al.

(1996) demonstrated that copepod grazing on microalgae enhanced the release o f metals

into the dissolved phase. Alternatively, S. alternijlora have been shown to take up metals

and then release dissolved metals through salt glands in the leaves (Burke et al. 2000).

Although salt glands appear to be more common in leaves than stems, it is possible that

dissolved metals were released directly from the stems o f the livingS. alternijlora , and

subsequently may have adsorbed to the epicuticle o f the grass shrimp during feeding

activities. Furthermore, grass shrimp constantly move up and down stems while scraping

epiphytes, and metal-enriched debris (epiphytes, exopolymers, etc) from the stems may

have adhered to the epicuticle o f the grass shrimp.

The route o f exposure significantly influenced partitioning o f Cu between the

exoskeleton and tissues inP. pugio. Cu and Zn bioaccumulation factors in the dietary

exposure experiment were higher in tissue than in exoskeleton. Fowleret al. (1970) also

found that more Zn was associated with tissues when uptake was from food due to direct

absorption across the gut wall and binding to tissue proteins associated with a regulatory'

process. More Cd was associated with the exoskeleton when the source o f exposure was

dietary (41% ) versus aqueous exposure (24%) (Table 3). Reinfelder and Fisher (1994)

hypothesized that the overwhelming association o f Cd with the exoskeleton o f

herbivorous copepods was accumulated from ingested food by rapid mobilization from

the gut and deposition in the exoskeleton as a means o f detoxification or storage.

Because seawater has a high concentration o f ions, interactions between solutes occur that

can alter the effective concentration (or activity) o f an ion (Libes 1992). Metals are more

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. likely in the dissolved state at low salinities, and are, therefore, more bioavailable and

toxic (Guerin and Stickle 1 9 9 5 ). However, Zn had higher bioconcentration factors at

30% o than at 5 and 18%o (Figure 6). Zn flux in crustaceans can vary with external Zn

concentrations, temperature, and between individuals (White and Rainbow 1 9 8 4 ). The

background body burdens o f the shrimp used in the 5 and 18 % o exposures were initially

high (Figure 4 ) . and it is possible that these high background burdens activated Zn

regulatory mechanisms. Zn is bound to hemolymph proteins, and excess Zn increases the

concentration o f unbound zinc in the hemolymph. and is usually removed rapidly by

excretion (Bryan 1 9 6 8 ). Another possible explanation for the apparent decrease in Zn

bioconcentration at low salinities could be due to a reduction in apparent water

permeability (APW) (Rainbow 1 9 9 5 ). At low salinities, many euryhaline crustaceans

decrease their AWP. causing a low uptake o f free ions. For example, the estaurine crab.

Carcinous maenas, from areas of high salinity exhibited lower uptake rates ofZn at lower

salinities than at high salinities due to changes in APW (Rainbow 1 9 9 5 ).

The results o f this study differentiate between superficially-adsorbed and internally

sequestered metals in the exoskeletonP. o pugio. f Surface-adsorbed metals were shown

to be an important part o f the body burden and their presence can lead to elevated body

burden values presumably without causing toxic responses. Exoskeleton

bioconcentration and bioaccumulation factors increased by 0-50% due to surface

adsorption to the epicuticle. Furthermore, salinity was shown to influence the surface

adsorption o f metals on the exoskeleton: low salinity reduced the amount o f adsorbed

metals. The exoskeleton P.o fpugio is not highly calcified nor waxy compared to other

decapod groups. Therefore, due to the differences in chemical composition of

5 6

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. exoskeletons, such as increase in calcification or waxes, there may be differences in

adsorption values for different crustacean taxa. Also, surface-area to body volume may

influence the relative importance o f adsorption, and adsorbed metals may be a higher

percentage o f the body burden o f very small crustaceans (larvae, meiofauna and

zooplankton). Future work may compare a broader range o f crustaceans from different

habitats. Nevertheless, the crustacean exoskeleton must be considered to understand

crustacean-metal interactions.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 4

TROPHIC TRANSFER OF CU, ZN, AND CD ASSOCIATED WITH EXOSKELETON AND TISSUES OF THE GRASS SHRIMP, PALAEMONETES PUGIO, TO A PREDATORY FISH, FUNDULUS GRANDIS

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. INTRODUCTION

Diet in fishes has been shown to be a predominant pathway for trace-metal

uptake, particularly Cd (Langston and Spence 1995; Watras and Bloom 1992).

Crustaceans, an important source o f prey for fishes, are exposed to heavy metals from

food, sediment and aqueous phases, and crustacean trace-metal body burdens are often

elevated, which may facilitate contaminant transfer via the food chain (Watras and Bloom

1992; King et al., 1992). However, the bioavailability o f metals to a predator depends on

the chemical form and binding within the prey's tissues as well as food quality (Langston

and Spence 1995).

Crustaceans all share the presence o f a chitinous exoskeleton, and metal cations

may associate with the chitin matrix by forming ionic bonds with nitrogen side groups

within the procuticle (Stevens 1985). Metals may thus be deposited within the

exoskeleton matrix associated with the deposition o f metal salts. Additionally, a Zn

metalloenzyme (Bryan 1968) and Cu-containing proteins (Twelliger 1999) are involved

in the tanning o f the exoskeleton. It was shown in Chapter 3 of this dissertation that a

significant portion o f the total metal-body burden may be associated with the chitinous

exoskeleton in the grass shrimp, and estimates o f the exoskeleton burden in other

crustaceans are reported as high as 97% o f the whole-body metal burden (Reinfelder and

Fisher 1994a). Low absorption efficiencies o f metals by the fishesMenidia menidia and

\t. beryllina fed copepod prey have been attributed to the proportion o f metals associated

with the exoskeleton and the indigestibility o f chitin (Reinfelder and Fisher 1994a).

High assimilation efficiencies in zooplanktivorous mysids were attributed to the selective

ingestion the soft tissues rather than exoskeleton (Smokerowskiet. al. 1998). However.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. most fish consume crustacean prey whole, and little is known about the bioavailibility o f

metals sequestered within the exoskeleton.

Metals may also associate with the arthropod exoskeleton through an adsorptive

process by binding externally to the surface o f epicuticle (Langston and Spence 1995;

Hare 1992). For example, it has been suggested that much o f the Zn associated with the

exoskeleton o f some decapod crustaceans is externally adsorbed because the metal

concentration o f the exoskeleton increases with increasing concentration in the ambient

water (Bryan 1968). Similarly. Munger and Hare (1997) suggested that Cd associated

w ith the exoskeleton o f a freshwater cladoceran was externally bound because Cd was

rapidly lost when transferred to Cd-free media. The adsorbed fraction o f the total metal

burden o (Palaemoneies pugio was measured at 18% o salinity in Chapter 3 o f this

dissertation, and averaged 17. 6. and 16% o f the total Cu. Zn and Cd burdens

respectively. Furthermore, adsorbed metals increased on the exuviae shed in metal-

enriched water as reported in Chapter 2. Many authors (Reinfelder and Fisher 1994a;

Munger and Hare 1997: Zanders and Rojas 1996: Smokerowski el al. 1998; Hare 1992)

acknowledge the presence o f surface-adsorbed metals in crustaceans and aquatic insects,

but the bioavailability o f this fraction to higher trophic levels has not been determined.

Because metals may be solublized in the gastric fluids o f predators, surface-adsorbed

metals may be an important source in the transfer o f metals to higher trophic levels.

The objective o f this portion o f my dissertation was to determine the role o f the

crustacean exoskeleton in the trophic transfer o f metals from a common and abundant

estuarine crustacean prey (Palaemoneies pugio Holthius) to a fish predator (Fundulus

grandis Baird and Girard). P. pugio is generally classified as a detritovore/ominvore and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. is a frequent prey o f larger fish species including species in the genusFundulus (Knieb

1987). P. pugio is also frequently used in toxicological studies (Klerks 1999; Wallace and

Lopez 1997; Smith and Weis 1997; Davis et al. 1998). Fishes in the genusFundulus are

important estuarine residents (Wainright el al. 2000; Rozas and Lasalle 1990; Kneib

1986) which prey upon invertebrates, and are also frequently used in toxicological studies

(Smith and Weis 1997: Davis et al. 1998: Vandenhurk et al. 1998). The assimilation of

metals adsorbed to the surface o f the grass shrimp exoskeleton, incorporated into the

chitin matrix, and associated with the soft tissues by F. grandis were compared.

METHODS

Approximately 800 P. pugio were collected from Spartinaa alternijlora salt

marsh near Cocodrie. La on May 6. 2000 (at 15 %o and 23° C) and June 11. 2000 (15 %o

and 29 o C). Half were subsequently exposed to an aqueous sublethal mixture o f Cd. Cu.

Zn (200 pg L"1 Cd. 1000 pg L '1 Zn. 500 pg L '1 Cu) in artificial seawater (ASW)

comprised o f ultra clean water (M illi-Q ) and Instant Ocean sea salts for three days in

acid-cleaned, low-density polyethylene (LDPE) 36-L plastic chambers. The remaining

shrimp were placed in ASW without amended metals. Chambers were aerated to

maintain high levels o f dissolved oxygen. Grass shrimp were selected on the basis o f size

(all were 22-28 mm in length from rostrum to telson) to maintain a consistent surface-

area- to-volume relationship and molt frequency, and shrimp were monitored daily for

molting during the course o f the exposure. Very few grass shrimp molted during the

exposure period (<10%).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Post Exposure

After 72 h. grass shrimp were removed from the tubs and frozen in liquid

nitrogen. H alf of the metal-exposed and unexposed grass shrimp were rinsed three times

for 20 min with 20 ml o f 10 mM ethylenediaminetetraacetate (EDTA) to chelate and

remove metals adsorbed to the surface o f the epicuticle. Exoskeletons were excised from

remaining tissues (excluding legs and antennae) o f the EDTA-rinsed (referred to as

"exoskeleton matrix‘*hereafter) and the unrinsed (referred to as "total exoskeleton"

hereafter) grass shrimp. The mass (wet weight) o f exoskeletons and tissues w'ere

determined with a Mettler balance (model number pm460. ±1 pg). Either total

exoskeleton, exoskeleton matrix, or tissues from three shrimp were mixed with 2-ml

Knox gelatin (following Wallace and Lopez 1997) and lg dissolved fish food (TetraMin).

The mixture was then placed in acid-cleaned small plastic conical molds to yield pellets

(containing the appropriate fraction o f three grass shrimp, fish food: approximate weight

was 2.5 g) palatable to fish.

Trophic Transfer

Fundulus grandis (ranging from 3- 4 g wet weight) were obtained from a bait

shop and maintained for ten days in 18 %o ASW. Five replicate 2.5-L glass fish bowls

were used to house single fish, in 18% o ASW. Fish were offered pellets containing either

the tissue, exoskeleton matrix, or total exoskeleton fraction o f grass shrimp either

exposed or unexposed to elevated metals. Each fish was fed a single pellet containing the

appropriate fraction o f three grass shrimp daily for 10 d. Unconsumed shrimp pellets

were removed daily and accounted for by mass. Water was replaced every other day in

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. all fish bowls to minimize ammonia and dissolved metal accumulation as a result o f

excretion or leaching from fecal pellets.

Samples (whole fish and three replicate subsamples o f each gelatin pellet

treatment) for metal analysis were placed in preweighed, acid-cleaned glass test tubes,

weighed, and digested in 5 ml o f concentrated trace-metal grade H N 03 (Fisher Scientific)

at 90° C for 12 h. The acid- containing digested samples were evaporated to about 1.5

ml. then diluted to -25 ml and reweighed to determine volume. Blanks were prepared for

each set o f samples digested. Metal concentrations in each fraction were determined by

Inductively Coupled Plasma (ICP) emission spectroscopy and corrected for interelement

interference.

For tissues and exoskeleton treatments, the mass o f each metal (per grass shrimp)

was estimated from the concentrations and total body mass. Metals adsorbed onto the

epicuticle surface were determined by calculating the difference in mass (per grass

shrimp) o f each metal in the unrinsed exoskeletons and exoskeletons rinsed with EDTA

o f both the metal-exposed and unexposed grass shrimp. Metal body burdens in F.

grcmiiis were measured and corrected for background body burdens. Assimilation

efficiencies (AEs) were also calculated based on the total mass o f each metal in fish tissue

and the total mass o f metal consumed by the fish as follows:

AE= Fish Tissue MetaKpg fish'1)* 100. Ingested Metals (pg fish 1)

Statistical Methods

The concentrations o f each metal in fish tissues were compared for each metal

with separate two-way ANOVAs in which main effects were exposure (metal-exposed or

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. unexposed shrimp) and grass shrimp fraction (total exoskeleton or tissues). AEs from the

two shrimp fractions (total exoskeleton or tissues) o f grass shrimp exposed or unexposed

to elevated metals were also compared for Cu and Zn using similar two-way AMOVAs.

Significant differences were analyzed with Tukey's post-ANOVA analysis. AEs for fish

fed grass shrimp not exposed to elevated Cd could not be calculated because

concentrations were below detection limits. Therefore. Cd AEs in fish fed metal-exposed

grass shrimp total exoskeleton and tissues were compared using a t-test. Concentrations

o ffish tissues and AEs o f the total exoskeleton (adsorbed + matrix) and exoskeleton

rinsed with EDTA (metals only in the matrix form) of shrimp exposed to elevated metals

were compared with t-tests to determine i f removing the adsorbed fraction influenced

trophic tranfer.

RESULTS

Shrimp Pellet Concentrations

Cu concentrations o f the food pellets containing exoskeleton matrix, total

exoskeleton, and tissues of grass shrimp not exposed to elevated metals averaged 19.1

(± 4.3 standard error o f the mean; this convention w ill be used in values that follow ). 21.2

(± 2.8). and 11.3 (± 0.8) pg g'1 wet weight. respectively (Figure 8). Based on the total

mass o f Cu associated with the fractions (3.1. 4.3. 5.1 pg. respectively). 13% was

adsorbed to the surface o f the exoskeleton. 33% was incorporated in the exoskeleton

matrix, and 54% o f Cu was associated with the tissues o f grass shrimp not exposed to

elevated metals (Table 5). Cu concentrations o f the pellets containing exoskeleton

matrix, total exoskeleton, and tissues o f grass shrimp exposed to elevated metals

averaged 20.0 (± 2.7). 34.9 (± 4.16). and 19.1 (± 0.7) pg g'1 wet weight, respectively

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 40

35 Cu 30

20 -I I | 15 1 I 10 I i 5 V s Z W k 11 0 II i Exposed Unexposed

- Zn 40 -i

Tissue 30 H Total Exo o> '•//yy' Exo Matrix O) ?////. a . 20

Y ////7 &

Exposed Unexposed

2.0

Cd 15

1.0 -I

0.5 -|

0.0

Exposed

Figure 8. Concentrations (pg gr1 wet weight) o f (a)Cu, (b) Zn. and (c)Cd in tissue, total exoskeleton, and exoskeleton matrix o f metal-exposed and unexposed grass shrimp. Palaemoneies pugio. N=3 and error bars represent standard error o f the meaa

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 5. Partitioning o f metals into exoskeleton matrix, surface-adsorbed, or tissues o f shrimp exposed or unexposed to Cu, Zn, and Cd expressed as a percentage o f the whole body burden. N.D.= no data, due to detection limits.

Exposed Unexposed Exposed Unexposed Exposed Unexposed Cu Cu Zn Zn Cd Cd

Matrix 22 33 30 36 52 N.D.

Adsorbed 23 13 8 14 7 N.D.

Tissue 55 54 61 54 41 N.D.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Figure 8). Based on the total mass o f Cu associated with the fractions (3.7, 7.6, 9.2 pg.

respectively). 23% o f grass shrimp Cu was adsorbed to the surface o f the exoskeleton.

22% was incorporated in the exoskeleton matrix, and 55% was associated with the tissues

(Table 5).

Food pellet Zn concentrations o f exoskeleton matrix, total exoskeleton and tissues

of grass shrimp not exposed to elevated metals averaged 24.5 (± 5.4). 29.2(± 4.0). and

15.6 (± 0.8) pg g'1 wet weight, respectively (Figure 8). Based on the total mass o f Zn

associated with the fractions (4.3. 6.0. 6.5 pg respectively) 14% was adsorbed to the

surface of the exoskeleton. 36% was incorporated in the exoskeleton matrix, and 54%

was associated with the tissues o f grass shrimp not exposed to elevated metals (Table 5).

Pellet Zn concentrations o f exoskeleton matrix, total exoskeleton and tissues o f grass

shrimp exposed to elevated metals averaged 36.3 (± 6.3). 39.8 (± 4.0). and 28.3 (± 2.1) pg

g’1 wet weight, respectively (Figure 8). Based on the total mass o f Zn associated with the

fractions (6.9. 8.7. 14.0 pg respectively). 8% o f the Zn mass associated with a grass

shrimp was adsorbed to the exoskeleton surface. 30% was incorporated into the

exoskeleton matrix, and 61 % was associated with the tissues (Table 5).

Concentrations o f Cd in pellets containing exoskeleton matrix, total exoskeleton

and tissues o f grass shrimp not exposed to elevated metals w ere below detection limits.

Cd concentrations in food pellets containing exoskeleton matrix, total exoskeleton and

tissues o f grass shrimp exposed to elevated metals averaged 0.90 (± 0.48). 1.10 (± 0.31).

and 0.50 (± 0.25) pg g'1 wet weight, respectively (Figure 8). Based on the total mass o f

Cd associated with the fractions (0.20. 0.23.0.25 pg. respectively), Cd adsorbed to the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. surface, incorporated in the matrix, and sequestered in the tissues represented. 7,41. and

54% respectively o f the total mass o f Cd per grass shrimp (Table 5).

Metal Concentration in Fish Tissue

Fish tissue metal concentrations were corrected for background body burdens in

F. grandis. Corrected-Cu concentrations in F. grandis tissues after feeding on shrimp

pellets for 10 d ranged from 0.01 to 0.16 pg g'1 wet weight across the three shrimp

fractions (exoskeleton matrix, total exoskeleton. or tissues) in fish that ingested shrimp

not exposed to elevated Cu. Corrected-Cu concentrations were typically higher (0.33-

0.49 pg g'1 wet weight across the three shrimp fractions) in fish that ingested shrimp

exposed to elevated metals (Figure 9). Rinsing with EDTA. to create an exoskeleton with

only matrix-bound Cu. did not reduce the mean fish tissue concentrations, suggesting

little or no uptake from the adsorbed fraction. Corrected-Zn concentrations in F. grandis

ranged from 0-25.0 pg g'1 wet weight across the three shrimp fractions in fish that

ingested shrimp not exposed to elevated Zn. Corrected-Zn concentrations were relatively

higher in fish that ingested shrimp exposed to elevated metals and ranged from 0.6-33.3

pg g’1 wet weight across the three shrimp fractions (Figure 9). Rinsing with EDTA

reduced corrected-Zn concentrations by about 90% compared to total exoskeleton.

suggesting a high uptake o f adsorbed Zn. Corrected-Cd concentrations in F. grandis

tissues ranged from 0.002-0.04 pg g'1 wet weight in fish that ingested shrimp not exposed

to elevated Cd. Corrected-Cd concentrations were typically higher (0.023-0.073 pg g'1

wet weight across the three shrimp fractions) in fish that ingested shrimp exposed to

elevated metals (Figure 9). Rinsing with EDTA reduced fish tissue concentrations by

about 70% suggesting a high uptake o f adsorbed Cd.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1.0

Cu 0.8

0 6

0.4 -I

0.2

0.0 Exposed Unexposed 40 -r

35 -j Zn i 30 j

25 -{ ~ ! Tissue i 77J2Z* Total Exo o > ; 20 i ESS3 Exo Matrix O) 15 -j

1 0 - j %

5 i 0 - Exposed Unexposed

0.20 Cd

0.15 -!

0.10 i

0.05 h

0.00 Exposed Unexposed

Figure 9. Funchdus grandis vvhole-body corrected concentrations (pg g-‘ wet weight) o f (a)Cu. (b) Zn. and (c) Cd fed either tissue, total exoskeleton. or exoskeleton matrix o f metal-exposed and unexposed P. pugio. N=5 and error bars represent the standard error o f the mean.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Two-way ANOVAs showed that fish-tissue metal concentrations were

significantly elevated for Zn and Cd when fed shrimp exposed to metals compared to fish

fed shrimp not exposed to elevated metals (Table 6). Shrimp fraction did not influence

fish tissue corrected concentrations for Cu and Cd. However, shrimp fraction did

influence the corrected Zn concentrations o f fish tissue, and fish fed shrimp tissue had

higher corrected concentrations o f Zn than fish fed total exoskeleton (p=<0.001).

Interactions were not significant for any o f the metals, t-tests showed that mean Cu. Zn

and Cd tissue corrected concentrations o f fish fed total exoskeleton and exoskeleton

matrix o f shrimp exposed to elevated metals were not significantly different (p=0.790.

0.080. and 0.221. respectively: Table 7).

Assimilation Efficiencies (AEs)

F. grandis Cu AEs after feeding on shrimp pellets for 10 days ranged from 0.13-

1.46% across the three fractions o f shrimp not exposed to elevated metals (Figure 10).

Cu AEs o f fish that ingested fractions o f shrimp exposed to elevated metals ranged from

1.2-5. %. The EDTA rinse did not decrease mean fish Cu AE. suggesting a very low

assimilation o f adsorbed Cu. Zn AEs ranged from 0.01 -133.4% in fish that ingested

shrimp fractions not exposed to elevated metals (Figure 10). Tissue AEs probably

exceeded 100% due to background variability among the shrimp or F. grandis. Zn AEs

offish which ingested fractions o f shrimp exposed to elevated metals ranged from 2.4-

94.6%. Highest AEs were always from the ingestion o f tissue. The EDTA rinse reduced

Zn AE by about 50%. suggesting a high assimilation o f adsorbed Zn. Because Cd

concentrations in shrimp were undetectable. AEs were calculated only for fish fed

fractions o f shrimp exposed to elevated metals, and values ranged from 6.7-12.2% across

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 6. Two-way ANOVA comparisons o f Cu, Zn, and Cd body burdens (pg g'1 wet weight) ofF. grandis fed P. pugio fractions (total exoskeleton and tissues) that were either exposed or unexposed to elevated metals. Values are P values from separate ANOVAs (n=6).

Metal Treatment (metal- Fraction (total exo, Interaction Term exposed vs. no tissues) metals)

Cu 0.069 0.611 0.835

Zn 0.018 <0.001 0.154 metal-exposed > no tissue > total exo metals

Cd 0.032 0.897 0.565 metal-exposed > no metals

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 7. t-test comparisons o f F. grandis concentrations o f fish that were fed either exoskeleton matrix (EDTA rinsed) or total exoskeletonP. o pugio f exposed to elevated metals (n=5).

Metal P value

Cu 0.790

Zn 0.080

Cd 0.221

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Exposed Unexposed Z m * 200 > t y 175 Zn C Q> 150 O ■ M B 125 3= _□ Tissue UJ 100 S3 Total Exo c 52332 Exo Matrix o 75 •** MB j 2 50 • MB 25 E Vs /S/ a vi m ’5> 0 I J7Z277]___ < Exposed Unexposed

25 -I Cd

10 1

Exposed

Figure 10.F. grandis (a)Cu, (b) Zn, and (c)Cd assimilation efficiencies fed either tissue, total exoskeleton. or exoskeleton matrix o f metal-exposed and unexposedP.pngio. N=S and error bars represent standard error o f the meaa

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the shrimp fractions (Figure 10). Rinsing reduced Cd AEs by about 30% suggesting that

some adsorbed Cd was assimilated.

Two-way ANOVAs showed that Cu and Zn F.grandis AEs were not significantly

different among fish fed shrimp exposed or not exposed to elevated metals (Table 8).

Shrimp fraction did not significanlty influence fish AEs o f Cu and Cd (Table 8). Zn AE.

however, was influenced by shrimp fraction (Table 8). Zn AEs were higher in fish fed

shrimp tissues than shrimp fed total exoskeleton. Interactions were not significant for

either metal. A t-test showed that Zn AEs were reduced by EDTA rinsing (Table 9). Cu

and Cd AEs from total exoskeleton and exoskeleton matrix o f shrimp exposed to elevated

metals were not significantly different (Table 9).

Percent Contribution

Using the total mass o f metals associated with a whole shrimp and the measured

mass o f metals assimilated from each fraction, the contribution (in %) o f a hypothetical

composite shrimp toF. grandis nietal-body burden was estimated (Table 10). Adsorbed

metals were estimated using the difference between total exoskeleton and exoskeleton

matrix values. According to these calculations, if a fish consumes a whole shrimp. 51%

and 49% o f the fish Cu burden could be attributed to the exoskeleton matrix and tissues

o f the shrimp, respectively. The surface-adsorbed fraction did not contribute to the fish

Cu burden. O f the Zn assimilated by F. grandis. 33. 2. and 65% could be attributed to the

adsorbed, exoskeleton matrix, and tissues o f the shrimp prey, respectively. For Cd. 38.

15. and 41% o f the fish burden could be attributed to the adsorbed, exoskeleton matrix

and tissues o f the shrimp respectively.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 8. Two-way ANOVA comparisons o f Cu and Zn, assimilation efficiencies o f F. grandis fed P. pugio fractions (total exoskeleton and tissues) that were either exposed or unexposed to elevated metals (n=5). t-test comparison o f Cd assimilation efficiency o f F.grandis fed P.pugio total exoskele.on or tissues (n=5). Values are P values from separate ANOVAs. N.D.= no data, due to detection limits.

Metal Treatment (metal - Fraction (total exo, Interaction exposed vs. no tissues) metals)

Cu 0.318 0.792 0.557

Zn 0.991 0.021 0.284 tissue > total exo

Cd N.D. 0.557 N.D

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 9. t-test comparisons o fF. grandis assimilation efficiencies that were fed either exoskeleton matrix (EDTA rinsed) or total exoskeletonP. o pugio f exposed to elevated metals (n=5).

Metal P value

Cu 0.841

Zn 0.008 total exo>exo matrix

Cd 0.300

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 10. Percent o f the metals in the exoskeleton matrix, surface-adsorbed, and tissues o fP. pugio estimated to contribute to the body burden o f F. grandis if whole shrimp were ingested.

Cu (%) Zn (%) Cd (%)

Matrix 51 2 15

Adsorbed 0 33 38

Tissue 49 65 41

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DISCUSSION

Metal uptake and assimilation efficiency by the predatory fish, F. grandis, from a

crustacean prey. P. pugio. varied among metal species and with the distribution o f metals

within the prey. These experiments utilized three shrimp fractions (tissues, total

exoskeleton, and exoskeleton matrix) fed separately toF. grandis for 10 days, and thus

directly compared metal uptake and assimilation from each fraction. Uptake by a

predator is a product o f the available metals from its prey and the efficiency by which

metals are assimilated. Grass-shrimp tissues were an important source in the uptake o f all

metals examined in F. grandis. Tissues comprised 55, 61. and 41% o f the Cu. Zn. and

Cd total prey body-burdens. respectively, and F. grandis AEs o f metals from exclusive

ingestion o f shrimp tissues tended to be as high or higher than the exoskeleton fractions.

Consequently. 49,65, and 41% o f the Cu, Zn. and Cd. respectively, accumulated by F.

grandis would be attributed to the tissues if whole P. pugio were ingested. Furthermore,

a distinction was made between metals adsorbed to the surface o f the exoskeleton versus

those sequestered within the exoskeleton matrix. The exoskeleton matrix did not

contribute substantially to the Zn (2%) or Cd (15%) burden o f F. grandis. Although the

matrix contained 30% o f the shrimp total Zn body burden. AE o f the matrix-bound Zn

was very low. The exoskeleton matrix comprised a relatively small fraction o f the total

shrimp body burden o f Cu (22%). but a high AE o f Cu from the exoskeleton matrix

(relative to the AE o f tissue Cu) suggested that the matrix would contribute 51% o f Cu

assimilated if whole grass shrimp were ingested. The AEs o f surface-adsorbed metals

ranged from 0-50%. Although the adsorbed proportion o f the shrimp total Zn and Cd

burden was relatively low (8 and 7%. respectively). F. grandis AEs from the adsorbed

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. fraction were high compared to the other fractions. The contributions o f the adsorbed

fraction were estimated to be 33% o f the total Zn and 38% o f the total Cd accumulated by

F. grandis if consuming whole shrimp. The surface-adsorbed Cu did not appear,

however, to contribute to the Cu burden oF. f grandis relative to the exoskeleton matrix

and tissues because AEs o f Cu were very low.

O f the three metals examined. Zn assimilation by F. grandis was most strongly

influenced by its distribution among shrimp exoskeleton and tissue fractions. Tissue and

surface-adsorbed Zn were highly bioavailable. whereas Zn bound to the exoskeleton

matrix was essentially not bioavailable to F. grandis. Previous authors have found that

metals dissolved in the cytosol o f tissues are more available to higher trophic levels than

metals precipitated in insoluble granules (Reinfelder and Fisher 1994b: Wallace and Lopez

1997: Wang et al. 1999). The observed high AE suggests that shrimp-tissue Zn is in a

soluble form, whereas the low' AE o f matrix Zn suggests that Zn is in an highly insoluble

form within the exoskeleton matrix of pugio.P. Zn adsorbed to the surface o f the

exoskeleton maybe is a surprisingly important pathway o f uptake by F. grandis. Surface-

adsorbed Zn was likely readily mobilized in the digestive tract o f F. grandis by the

digestive juices and absorbed across the intestinal wall (Langston and Spence. 1995).

Wang et al. (1999) found that Zn associated with the soft tissues o f copepods did not

influence uptake by barnacles and suggested that Zn associated with the exoskeleton was

assimilated by barnacles. In contrast. Reinfelder and Fisher (1994a) found a correlation

between Zn associated with the soft tissues o f copepod and assimilation by fishes in the

genus Menidia. However, these correlations were indirectly inferred from the AEs and the

proportion associated with the soft tissues o f the copepod prey and were not based on

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. direct measurements o f assimilation o f separate fractions. Differences in digestive

physiology o f the predator, metal solubility in the prey tissues and exoskeleton, and

assimilation o f the surface-adsorbed fraction may lead to differences in reported metal

AEs.

An unexpected result o f this study was the high AE o f Cu from the exoskeleton

matrix (exoskeleton matrix AEs were higher than from the tissues). Although the

exoskeleton is not generally subject to digestion, the binding o f metals to relatively

digestible proteins in the exoskeleton procuticle may control Cu bioavailability. Metals

associated with soluble proteins are available to higher trophic levels (Reinfelder and

Fisher 1991) and may be present in the exoskeleton. Phenoloxidases. Cu-containing

proteins similar to the hemolymph proteins in the tissues (hemocyanin), are also present in

the crustacean exoskeleton ( Terrwilliger 1999). Therefore, the similarity in Cu AEs o f the

exoskeleton and tissue may be the result o f the similarity in digestibility o f the Cu-

containing proteins associated with the tissues and exoskeleton. To my knowledge, no

studies have been conducted to examine the assimilation o f Cu from prey hemocyanin by

consumers. My calculations estimate that Cu associated with the matrix o f the

exoskeleton and tissues contribute about equally to the total Cu body burden o f F. grandis

(51% and 49% respectively. Table 10). The surface-adsorbed Cu does not appear to

contribute to F. grandis Cu burden relative to the exoskeleton matrix and tissues. It is not

clear if this result is due to a different mechanism for binding surface-adsorbed Cu

compared to Zn and Cd. or if the fate o f surface-adsorbed Cu is different from other

sources after assimilation.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. My measurements o f Cd AEs in F. grandis (7-20%) were slightly lower than those

reported by Ni et al. (2000) for mudskipper. Periophthalmus cantonensis, and glassy.

Ambassis urotaenia. fishes (10-33%). yet slightly higher than those reported by Reinfelder

and Fisher (1994a) for silversides (2.7%). Furthermore, the mean AEs by F.grandis o f Cd

in the total exoskeleton and tissues were not distinguishable statistically. Ni et al. (2000)

did not find a correlation between assimilation of Cd by the glassy and the distribution

between exoskeleton and tissue o f zooplanktonic prey. Nevertheless, a significant

correlation between Cd assimilation and metal distribution in zooplanktonic prey tissues

was found for the mudskipper (Ni et al. 2000). and Reinfelder and Fisher (1994) found a

correlation between the percent o f the metals in the tissues o f copepod prey and the AEs o f

Cd by silversides. Although these correlations are based upon indirect evidence, the

assimilation o f exoskeleton-bound metals may vary because o f differences in digestion

processes among fishes(i.e.. differences in gut pH. gut passage time). Furthermore, the

surface-adsorbed Cd was highly available to F. grandis. The 15% contribution of the

adsorbed fraction suggests that surface-adsorbed Cd is desorbed in the gut o f F.grandis

(see alsoLangston and Spence 1995). Variation in the adsorbed fraction may account for

the variability reported in the assimilation o f Cd associated with the exoskeleton (Ni et al.

2000: Reinfelder and Fisher 1994a). and is generally a neglected component o f trophic

transfer.

Fishes generally have been found to regulate essential elements such as Cu and Zn

(Langston and Spence 1995). For example. Cu concentrations o f Pleuronectes platessa

remained unchanged regardless o f exposure concentrations over a range o f sublethal

conditions (Saward et al. 1975). In addition to physiological regulation, low body burdens

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. may be due to low uptake because dietary Cu is likely complexed by mucus present in the

gut wall and egested (Handy 1986). AEs o f F.grandis for Cu, Zn, and Cd were /ery

different from each other, with highest AEs for Zn and lowest for Cu. The generally low

assimilation efficiencies (AE < 2%) of Cu by F. grandis may be the result o f physiological

regulation or low adsorption across the gut wall. The Zn uptake pathway is similar to Cu,

and may result in competition for binding sites, which could possibly account for the

difference in AE o f the two metals. Zn AE o f 70% are considered typical for fishes, and

Zn AE in the present study was similarly high (from 50-100%). but varied among shrimp

fractions, suggesting that excess Zn was not regulated by F. grandis efficiently as was Cu.

or that this species may have high metabolic requirements o f Zn. Wang et al. (1999) also

reported high Zn AEs in barnacles (93%) ingesting copepod prey. Other authors have

reported that Zn is more bioavailable to fishes than other elements including Cu and Cd

(Reinfelder and Fisher 1994a; Cross et al. 1975). These observations suggest that there

may be interactions between metal species by processes such as substitution and

competition for binding sites between metals, which may lim it uptake (Zyadah and Abdel-

Baky 2000). Zn has been shown to reduce Cu body burdens in rainbow trout (Mount et al.

1994). and Zn may also inhibit Cd uptake (Goyer 1995). Therefore, the high Zn body

burdens may have reduced the assimilation efficiencies o f Cu and Cd by F. grandis.

Variability o f metal uptake could be due to physiological regulation and could

result in the under or over estimation o f the true proportion o f metals assimilated byF.

grandis. An alternative approach to estimate metal uptake would be to measure

absorption efficiency (the fraction o f the ingested metal that is taken up across the gut

wall. Penry 1998). Absorption efficiency is generally measured using the portion o f the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. metals egested in short-term, pulse-chase experiments. Using absorption efficiencies in

concert with the assimilation efficiencies should provide a more accurate estimation o f

bioaccumulation from the different fractions o f P. pugio. In this study F. grandis feces

were collected every other day. and subsequently analyzed for metal content in an attempt

to measure absorption efficiency. However, metal concentrations in the feces were

extremely high, probably due to the scavenging o f metals that were excreted into the

water. In order to obtain an accurate estimate o f adsorption efficiencies, radionuclides

utilized in pulse-chase experiments would be useful; however the short half life of wCu

(half-life = 12.7 h) makes radionuclide work with Cu generally problematic. Furthermore,

this study was designed to mimic natural conditions as closely as possible, and the use o f

radiotracers may cause artifacts such as differences in biogeochemical behavior of the

radioisotopes from the naturally occurring stable isotopes. Paulson and Gendron (2001)

found that the association o f 64Cu with organic particles was 70% higher than that o f

natural Cu. suggesting that measurements o f assimilation efficiency using h4Cu may

overestimated true rates o f uptake.

Biota are known to be an important influence on the cycling o f metals in the

environment through mechanisms such as sequestration and trophic transfer (Wang and

Fisher 1995). Because crustaceans are abundant and widespread, it is important to

understand how trace-metal distributions among tissues and exoskeletons o f crustacean

prey influence transfer to higher trophic levels. The findings o f this study suggest that the

bioavailability o f metals associated with the tissues and exoskeleton (both incorporated

within the chitin matrix and surface-adsorbed) o f P. pugio to F. grandis varied among

metals. Zn associated with the tissues was shown to be highly available to F. grandis.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. whereas there was no ditYerence in AE o f Cu and Cd associated with either tissues or total

exoskeleton. Furthermore. Cd and Zn adsorbed to the surface o f shrimp exoskeleton

potentially contribute a high fraction o f the total metal uptake o f F. grandis from grass

shrimp. The assimilation o f metals associated with the different fractions o f the grass

shrimp may be a function o f chemical speciation (solubility) o f the metals and the

digestive physiology o f the predator. Variation among crustacean prey exoskeleton and

size (surface-area-to-body-volume ratio) may result in variation in the surface-adsorbed

fraction and therefore may contribute to the large variability among metal AE reported in

other studies (Reinfelder and Fisher 1994a: Ni et al 2000: Wang et al 1999; Smokerowski

et al. 1998) that have examined the uptake o f metals from crustacean prey. It is often

presumed that metals associated with the exoskeleton are unavailable to higher trophic

levels; however, the results o f this study indicate that the adsorbed exoskeleton-associated

metals can be important an vector in the transfer o f metals to higher trophic levels.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 5

SUMMARY AND CONCLUSIONS

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Although bioaccumuiation o f trace metals by crustaceans has been investigated in

numerous studies, relatively little research has been done to clarity the role o f the

exoskeleton in the distribution and sequestration o f trace metals, or to determine how

various abiotic and biotic factors influence this role. The limited studies that have

examined the association o f metals with the crustacean exoskeleton report varying results,

and a major goal o f this dissertation was to determine if molt cycle, salinity, and route o f

exposure explain the discrepancies in the reported values. Furthermore, although many

authors acknow ledge the presence o f surface-adsorbed metals on the crustacean epicuticle

(Munger and Hare 1997: Reinfelder and Fisher 1994b: Zanders and Rojas 1996). to my

knowledge no researchers have made quantitative measurements o f the contribution o f

this fraction to the whole-body burden in any crustacean. Therefore, in my dissertation

the fraction of the whole-body burden that was associated with the tissues, adsorbed to

the surface o f the exoskeleton. and associated with the matrix o f the exoskeleton were

each quantified. The exoskeleton fractions were shown to vary under different

conditions. Finally, the ecological relevance o f the partitioning between exoskeleton and

tissues was demonstrated by examination o f trophic transfer to a predatory fish. It has

often been presumed that exoskeleton-associated metals are unavailable to predators due

to the indigestibility o f chitin. A few studies have attempted to determine indirect

correlations between metal assimilation efficiencies in predators and the proportion o f

total body burden associated with soft tissues o f crustacean prey, but no direct

measurement o f trophic- transfer efficiency has yet been published.

Results o f the second chapter o f this dissertation suggested that molt-cycle stage is

important to the fate and distribution o f Cu. Zn. and Cd in P. pugio. Furthermore, the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. fate o f exoskeleton-associated metals varied for each metal species examined. Most o f

the whole-body-Cu burden appeared to be retained during a single molt; exoskeleton-

associated Cu may therefore be reabsorbed before ecdysis. This conservation o f Cu may

be associated with the presence o f Cu-containing proteins, prophenoloxidases. that cross

link and harden the exoskeleton during ecdysis. Total Cd body burden was significantly

reduced by a single molt; however, some fraction o f the Cd burden in the exoskeleton

was likely reabsorbed into the tissues. Some Cd was probably mobilized and transported

to the tissues during molting because a relatively low fraction (25%) o f the total burden

was lost with a single molt. If such mobilization and transport is common within many

crustaceans, the redistribution o f Cd into the tissues may be the cause o f the frequently-

observed increase in the mortality o f molting crustaceans. Finally, my data demonstrated

that post-ecdysis Zn whole-body burdens were significantly lower than the burdens o f

intermolt grass shrimp, but relatively little Zn was associated with the exuviae.

Therefore. Zn appeared to be depurated by P. pugio through a physiological mechanism

other than ecdysis.

Another finding presented in Chapter 2 was that metals adsorbed onto the surface

o f exuviae shed in metal-enriched w ater. A substantial quantity o f metals adsorbed onto

the exuviae are likely due to an increase in available binding sites caused by the exposure

o f the underside o f the molted exoskeleton. Because many crustaceans consume their

exuviae, elevated metals associated with exuvial adsorption may be an important pathway

o f uptake in crustaceans.

These findings may have important implications with regard to metal fate in

crustaceans. Metals may not exert toxic effects while retained within the exoskeleton and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in fact, it is often presumed that metals are stored in the exoskeleton for potential

depuration through ecdysis. However, this study demonstrated that there is variation in

the distribution o f different metal species before and after ecdysis. Furthermore, it

appears that some fraction o f the total metal burden is readsorbed before ecdysis and is

not depurated. If metals are redistributed before ecdysis, tissue concentrations may vary

widely over a molt cycle. Metals reabsorbed into the tissues may cause a body burden

that exceeds critical body residues, leading to consequent mortality. This sudden increase

in tissue concentrations at specific times in the molt cycle may reduce the usefulness o f

crustaceans as test organisms in toxicological bioassays because their pattern o f toxicity

may not be representative of non-molting animals. Future work should be done to further

determine the fate o f toxic metals during ecdysis and if (and by how much) tissue burdens

increase as a result o f redistribution from the exoskeleton. The correlation o f critical

body residues and tissue-reabsorption may explain why there is an apparent increase in

mortality among crustaceans associated with ecdysis.

The results presented in Chapter 3 o f this dissertation differentiated between

surface-adsorbed metals and metals internally sequestered in the matrix o f the

exoskeleton. Surface-adsorbed metals led to increases in whole-body burden values and

an apparent increase (0-50%) in metal bioconcentration and bioaccumulation factors,

presumably without causing toxic responses. Adsorbed Cu. Zn. and Cd were as high as

20. 18. and 25%. respectively o f the whole-body burden, in P. pugio. Therefore,

estimated crustacean body burdens that induce mortality (the so called "critical body

residues” ) may be inflated if the absorbed metals are included in the body burden

measurement. Salinity was shown to influence the adsorption o f metals onto the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. exoskeleton (at 5%o only 0-4% o f the whole-body burden o f Cu, Zn. and Cd was due to

the adsorbed fraction). Additionally, differences in the chemical composition, such as the

degree o f calcification and the amount o f waxes on the exoskeleton. may lead to

differences in surface adsorption among different crustacean taxa. Furthermore, surface-

area-to-body volume ratio may influence the relative importance o f adsorption, and the

adsorbed metals may represent a higher portion o f the total-body burden in smaller

crustaceans, particularly larvae, zooplankton and meiofauna. These factors may

contribute to the variation o f the apparent bioaccumulation o f metals among crustaceans.

Future work should include the measurement o f the adsorbed fraction across different

taxa o f varying size classes with different exoskeleton chemical compositions to more

broadly characterize its importance.

Metals associated with the exoskeleton matrix also varied with each metal

examined and ranged from 7-45% o f the whole-body burden. Zn and Cd levels were

particularly high within the exoskeleton matrix, perhaps due to their chemical similarity

with Ca. Cu and Zn tissue burdens showed very little variability among dietary and

aqueous exposure treatments, presumably because they are physiologically regulated. Cd

tissue burdens varied across different treatments possibly because Cd has no known

metabolic function and is not regulated. The chemical form o f stored metals associated

with the exoskeleton matrix is currently unknown. Although meta'-detoxitying granules

are reported in the digestive tissue o f crustaceans, to my knowledge there are no reports

o f metal-containing granules in the crustacean exoskeleton with the exception o f Ca

(Greenaway and Farrelly 1991). Therefore, it is possible that the substitution o f metals

with Ca could result in the production o f insoluble granules rich in Cd and Zn.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Knowledge o f the chemical form o f metals is necessary to explain toxic effects to the

organism because insoluble granules render metals nontoxic. However, Cu in the matrix

may have been associated with proteins. More work should be done to determine the

distribution and chemical behavior o f metals within the procuticle and chitin matrix.

The accumulation of metals in the tissues, exoskeleton matrix, and total

exoskeleton oP. f pugio from aqueous (as a function o f salinity) and dietary exposure

varied for each metal examined. Zn was chosen here as a graphic benchmark with which

to compare the other metals examined because o f its prevalence in the literature and it is a

regulated, essential metal (Brian 1 9 6 8 ; White and Rainbow 1 9 8 4 ; Fowleret al. 1 9 7 1 ;

Wang et al. 1 9 9 9 ). Bioconcentration o f Cd in the exoskeleton fractions (matrix and total

exoskeleton) always exceeded that o f Zn regardless o f salinity (Figure 11). Cu and Cd

bioconcentration in the tissues and exoskeleton (matrix and total exoskeleton) were

higher than that o f Zn at 3 0 % o . Zn bioaccumulation and bioconcentration at 18% o were

also compared with the other metals, again as an important benchmark. Cu and Cd

bioaccumulation factors were always lower than bioaccumlation factors (Figure 12).

However. Cu bioconcentration factors were consistently higher than Zn bioconcentration

factors for tissues, total exoskeleton. and exoskeleton matrix. Cd bioconcentration

exceeded Zn only in the tissues. This variation in the uptake and partitioning o f each

metal may influence critical body residues and the toxicity o f mixtures. Future work

should compare the accumulation o f additional metals in the exoskeleton and tissues o f P.

pugio and other crustaceans.

Chapter 4 addressed the ecological relevance o f the partitioning o f metals between

the exoskeleton and tissues inP. pugio. The metal distribution between tissues and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 150

100

Zn (Benchmark) N UL CO to

-100 5 ppt 18 ppt 30 ppt

Cu Tissue Salinity o Cd Tissue Cu Total Exo —v Cd Total Exo Cu Matrix —□ • Cd Matrix

Figure 11.The differences in Cu and Cd bioconcentration factors using Zn as a benchmark (solid line) o f tissues, total exoskeleton. and exoskeleton matrix for Palaemomtes pugio plotted across salinity exposures. The v axis represents the difference between Zn and Cu bioconcentration factors and Zn and Cd bioconcentration factos

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced Palciemonetes pugio. Figure 12. Figure Bioaccumulation Factor Bioconcentration Factor factors in the tissues, total exoskeleton, and exoskeleton matrix o f f o matrix andexoskeleton exoskeleton, thetissues, total in factors 0 0 1 • 0 ▼ Zn Cu Cd Cu. Zn (benchmark, solid line), and Cd bioaccumulation and bioconcentration bioconcentration and bioaccumulation andCd line), solid (benchmark, Zn Cu. Tissue Tissue Q Zn (Benchmark) Zn Q Zn (Benchmark)Zn Total Exo Exo Matrix Exo Exo Total oa Eo x Matrix Exo Exo Total 92 exoskeleton was found to influence trophic transfer to the predatory fishFundulus

grandis. The assimilation efficiencies by F. grandis o f metals adsorbed to the

exoskeleton surface, accumulated within the matrix, or associated with the tissues varied

among metal species. It is often presumed that metals associated with the exoskeleton are

unavailable to higher trophic levels, but the results o f Chapter 4 indicated that

exoskeleton-associated metals, particularly the surface-adsorbed metals, can be an

important vector in trophic transfer to a predator that consumes whole crustacean prey. F.

grandis assimilation efficiencies from the different fractions varied for each metal. The

assimilation efficiencies o f Cu and Cd associated with the exoskeleton matrix were not

different from the tissues of P.pugio and ranged from 1-20% . If F. grandis were to

consume whole shrimp, approximately 51 and 15% o f the Cu and Cd. respectively,

accumulated could be attributed to the exoskeleton matrix; whereas. 49 and 41%.

respectively, was estimated to be contributed by shrimp tissues. Furthermore. Zn

associated with the grass shrimp tissues was highly available (near 100%). while Zn

sequestered in the exoskeleton matrix was essentially unavailable to F. grandis.

Hypothetically. 65% o f the Zn accumulated by F. grandis was associated with the tissues

o f whole shrimp, yet only 2% was estimated to be accumulated from the exoskeletcn

matrix. This study also concluded that Zn and Cd adsorbed to the surface o f the

exoskeleton are highly bioavailable to F. grandis. Thirty-three percent o f the Zn and 38%

o f the Cd accumulated by F. grandis from hypothetical whole shrimp was estimated to be

adsorbed to the exoskeleton surface.

The assimilation o f Cu. Zn. and Cd from the different shrimp fractions by F.

grandis varied greatly for each metal. Zn was again chosen as a benchmark to compare

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. AE with the other metals studied due to its prevalence in the literature and it is regulated.

Assimilation efficiencies o f Cu and Cd associated with the exoskeleton matrix were

slightly higher than the assimilation o f matrix Zn. although all matrix AEs were generally

low (Figure 13). Zn AEs were much higher than Cd and Cu for tissues and total

exoskeleton. Although surface-adsorbed Zn was highly bioavailable to F. grandis,

adsorbed Cu was probably not readily assimilated. Therefore, the mixture of metals may

contribute the variation o f assimilation efficiencies from the tissues, total exoskeleton and

exoskeleton matrix. The assimilation o f other metals associated with the tissues and

exoskeleton fractions should be compared in future investigations with crustaceans.

Metal surface-adsorption also varied with salinity and for each metal. Surface-

adsorption also may vary due to differences in exoskeleton composition, surface-area-to-

body volume ratio, and exposure time. Such differences likely lead to variations in

assimilation efficiency by predators consuming crustacean prey. Therefore, in order to

predict trace-metal fate i:i aquatic ecosystems through trophic transfer to predators that

consume whole crustacean prey, an improved knowledge o f surface-adsorbed metals is

required. More studies are needed with different types o f crustacean prey because

exoskeleton composition varies and partitioning with the exoskeleton may also vary.

Other metals may have a different affinity with the exoskeleton than those examined here,

and should be compared in order to determine the bioavailability to predators. This study

was conducted on one species o f fish, and other fishes may have physiological differences

(i.e.. gut passage time, gut pH) that w ill influence the assimilation o f metals from

crustacean prey. The assimilation o f exoskeleton-associated metals by different species

o f fishes would provide additional insight. Future studies may utilize radiotracers in

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Zn (Benchmark) 100 -

75 -

Ui 50 - <

25 -I

_Q_ O Tissue Total Exo Exo Matrix

• Cd O Cu

Figure 13. Fimdulus grandis assimilation efficiencies o f Cd and Cu from tissues, total exoskeleton, and exoskeleton matrix oPalaemonetes f pngio compared to Zn (solid line) as a benchmark.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. order to measure adsorption efficiencies and a thorough mass balance study would further

elucidate metal fates. The assimilation o f metals associated with the exoskeleton and

tissues may be a function o f solubility o f metals. X-ray emission analysis o f exoskeletons

o f crustaceans exposed to elevated metals would be useful to determine if various metals

are stored in insoluble granules in the exoskeleton, and may be necessary to explain

differences in assimilation efficiencies o f the three metals studied.

The results o f this dissertation have important implications with regard to how

investigations using crustaceans exposed to trace metals should be conducted. The

presence o f surface-bound metals associated with the exoskeleton suggest that whole-body

burden estimates that do not first remove such metals are greater than metal burdens that

are actually experienced in the tissues. Thus, estimates o f body burdens which cause

mortality, such critical-body-residues. may be higher than what tissues actually experience

and may be inflated. Surface-adsorbed metals may also lead to variability in

biomonitoring data based on whole-body burdens due to different degrees o f surface-

adsorption on crustaceans o f different sizes and from different habitats o f varying salinity.

Therefore, in order to obtain accurate correlations between tissue residues and toxic

effects, results o f this dissertation suggest that crustaceans should be rinsed with a strong

chelator such as EDTA prior to analysis to remove surface-adsorbed metals. On the

contrary, absorbed metals contribute a large fraction o f the trophic transfer o f metals to

predators. Investigations designed to model the trophic transfer o f metals from crustacean

prey should include the surface-adsorbed metals as part o f the whole-body burden. Thus,

the goals o f the project o f interest involving crustaceans and metal-body burdens should be

considered before a specific technique is chosen.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LITERATURE CITED

Adams, E., Simkiss, K., and Taylor, M. (1982) Metal ion metabolism in the moulting crayfish (Austorpotamohius pallipes). Comp. Biochem. 72A: Physiol. 73-76.

Bertine, K.K. and Goldberg, E.D. (1972) Trace elements in clams, mussels, and shrimps. Limnol. Oceanog. 17: 877-884.

Brown. B.E. (1982) The form and function o f metal-containing granules in invertebrate tissues. Biological Reviews. 57: 621-667.

Bryan. G.W. (1968) Concentrations o f zinc and copper in the tissues o f decapod crustaceans../ A far. Biol. Ass. U.K. 48: 303-321.

Burke. D.. Weis. J.S.. Weis. P. (2000) Release o f metals by the leaves o f the salt marsh grasses Spariina alterniflora and Phragmites australis. Estuarine Coastal ami Shelf Science .51: 153-159.

Catallo. W.J.. Henk. W.. Younger. L.. M ills. O. Thiele. D.J.. and Meyers. S.P. (1996) Trace metal uptake by Pichia Spartinae. an endosymbiotic yeast in the salt marsh cord grass Spartina alterniflora. Chemistry ami Ecology'. 13: 113-131.

Chen. C.Y.. Stemberger.R.S.. Klaue. B.. Blum. J.D.. Pickhardt. P.C., and Folt. C.L. (2000) Accumulation o f heavy metals in food web components across a gradient o f lakes. Limnol. Oceanogr. 45: 1525-1536.

Cross. F.A.. Flardy. L.H.. Jones. N.Y. and Barber. R. (1973) Relation between total body weight and concnetrations o f manganese, iron, copper, zinc and mercury in white muscle o f bluefish (Pomatomus saltatrix) and a bathyl-demersal fish Antimora rostrala. J.Fish. Res. Bd. Can. 30: 1287-1298.

Depledge. M.H.. Forbes. T.L.. and Forbes. V.E. (1993) Evaluation o f cadmium, copper, zinc, and iron concentrations and tissue distributions in the benthic crab. Dorippe granulata (De Haan. 1841) from Tolo Harbour. Hong Kong. Environmental Pollution 81. 15-19.

Engle. D.W. (1987) Metal regulation and molting in the blue crab, Callinectes sapidus: copper, zinc, and metallothionein. Biol. Bull. 172: 69-82.

Fisher. W.S. and Foss. S.S. (1993) A simple test for toxicity o f number 2 fuel oil and oil dispersants to embryos o f grass shrimp, Palaemonetes pugio. Mar. Poll. 26: Bull. 385-391.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fleeger, J.W., Carman. K.R., Webb. S., Hilbun .N.. & Pace. M.C. (1999) Consumption o f microalgae by the grass shrimp. Palaemonetes pugio. Journal o f Crustacean Biology 19: 324-336.

Fowler. S.W.. Small, L.F.. and Dean. J.M. (1970) Distribution o f ingested zinc-65 in the tissues o f some marine crustaceans. Journal Fisheries Research Board of Canada 27: 1051-1058.

Fowler. S.W., Small. L.F.. and Dean. J.M. (1971) Experimental studies on elimination o f zinc-65. cesium-137and cerium-144 by euphasusiids. Mar. Biol. 8: 224-231.

Greenaway. P. and Farrelly. C. (1991) Trans-epidermal transport and storage o f calcium in hothusiana-transversa (brachyura. sundathelphusidae) during premoult. Acta Zoologica 72 (1): 29-30.

Guerin. J.L. & Stickle. W.B. (1995) Effects o f cadmium on survival, osmoregulatory ability and bioenergetics o f juvenile blue crabs Callinectes sapidus at different salinities. Mar. Environ. Res 40: 227-246.

Guillarume.J. (1997) Proteins and Amino Acids. In Crustacean Nutrition. (Eds. L.R. D’Abramo. D.E. Conklin, and D.M. Akiyama). pp. 28-29. World Aquaculture Society. Baton Rouge. LA.

Hagerman. L. (1983) Haemocyanin concentration o f juvenile lobster (Homarus gammarus) in relation to moulting cycle and feeding conditions. Mar. Biol. 77: 11- \ l .

Hebei. D.K.. Jones. M.B. & Depledge. M.H. (1997) Responses o f crustaceans to contaminant exposure: A holistic approach. Estuar.Coast.Shelf.Sci. 44: 177-184.

Handy. R.D. (1996) Dietary exposure to toxic metals in fish. In: Toxicology o f Aquatic Pollution: Physiological. Cellular and Molecular Approaches. (Ed E.W.Taylor), pp. 29-60.

Hare. L. (1992) Aquatic insects and trace metals: bioavailability, bioaccamulation. and toxicity. Critical Reviews in Toxicology 22(5/6): 327-369.

Keteles. K.A. and Fleeger. J.W. (In Press.) The contribution o f ecdysis to the fate o f copper, zinc and cadmium in grass shrimp. Palaemonetes Pugio Holthius. Mar. Poll. Bull.

Key. P.B. and Fulton. M.H. (1993) Lethal and sublethal effects o f chlorpyrifos exposure on adult and larval stages o f the grass shrimp Palaemonetes pugio. J. Environ. Sci. Health 28: 621-640.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Khan, A.T., Weis, J.S.. D'Andrea, L. (1989) Bioaccumulation o f four heavy metals in two populations o f grass shrimp. Palaemonetes pugio. Bull. Environ. Contam. Toxicol.42: 339-343.

Klerks, P.L. (1999) Acclimation to Contaminants by the Grass Shrimp Palaemonetes pugio: Individual Contaminants vs. Mixtures. Ecotoxicology 8: 277-286.

Kneib. R.T. (1987) Predation risk and use o f intertidal habitats by young fishes and shrimp. Ecology 68: 379-386.

Langston. W.J. & Spence, S.K. (1995) Biological Factors Involved in Metal Concentrations Observed in Aquatic Organisms. In: Metal Speciation and Bioavailability in Aquatic Systems (Eds. Tessier. A. & Turner, D.R.), pp. 407-478. John Wiley & Sons Ltd.

Libes, S. (1992) Introduction to Biogeomarine Chemistry. John Wiley and Sons Ltd.

Morgan. M.D. (1980) Grazing and predation o f the grass shrimp Palaemonetes pugio. Limnol. Oceanogr.. 25: 896-902.

Mount. D.R.. Barth. A.K. Garrison. T.D. Barten, K.A. and Hockett. J.R. (1994) Dietary and waterborne exposure o f rainbow trout ( Oncohynuchus mykiss) to copper, cadmium, lead and zinc using a live diet. Environmental Toxicology' and Chemistry 12: 2031 -2041.

Munger, C. and Hare. L. (1997) Relative importance o f water and food as Cd sources to an aquatic insect (Chaoborus punctipennis): implications for predicting Cd bioaccumulation in nature. Environmental Science and Technology. 31: 891-895.

Ni. I.H. Wang. W.X.. and Tam. Y.K. (2000) Tranfer o f Cd. Cr, and Zn from zooplankton prey to mudskipper Periophthalmus cantonensis and glassy Ambassis urotaenia fishes. Mar. Ecol. Prog Ser. 194: 203-210.

Nieland. D.L. Wilson. C.A.. Fleeger. J.W. Sun. B. Malone. R.F. and Chen. S. (1998) Preliminary evaluation o f the use o f phosphogypsum for reef substrate. I. A laboratory study o f bioaccumulation o f radium an six heavy metals in an aquatic food chain. Chemistry and Ecology 14: 305-319.

Nimrno. D.W.R.. Lightner. D.V.. and Bahner. L.H. (1977) Effects o f Cadmium on the Shrimps. Penaeus duorarum. Palaemonetes. and pugio Palaemonetes vulgaris. In: Physiological REsponses o f Marine Biota to Pollutants. (Eds.Vemberg. F.J.. Thurberg.. F.P.. Calabrese. A. and Vernberg. W). pp. 131-184. Academic Press. NY.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Paulson, A.J. and Gendron, J.F. (2001) Partitioning o f copper at concentrations below the marine water quality criteria. Environmental Toxicology and Chemistry 20: 952- 959.

Penry. D.L. (1998) Applications o f Efficiency Measurements in Bioaccumulation Studies: Definitions. Clarifications, and a Critique o f Methods. Environmental Toxicology’ and Chemistry 17: 1633-1639.

Russell. R.W.. Gobas P.C., and Haffner.G.D. (1999) Role o f chemical and ecological factors in trophic transfer o f organic chemicals in aquatic food webs. Environmental Toxicology and Chemistry 18: 1250-1257.

Scott-Fordsman. J.J. and Depledge. M.H. (1997) Changes in the tissue concentrations and contents o f calcium, copper and zinc in the shore crab Carcius maenas (L.) (Crusatacea: Decapoda) during the moult cycle and following copper exposure during ecdysis. Marine Environmental Research 44: 397-414.

Smokorowski. K.E.. Lasenby. D.C. and Evans. R.D. (1998) Quantifying the uptake and release o f cadmium and copper by the opossum shrimp Mysis relicta preying upon the cladoceran Daphnia magna using stable isotope tracers. Can. J. Fisheries Aquat. Sci. 55: 909-916.

Rainbow P.S. (1995) Physiology. Physicochemistry and Metal Uptake—A Crustacean Perspective.Mar. Poll. Bull. 31: 55-59

Reinfelder. J.R. and Fisher. N.S. (1991) The assimilation o f elements ingested by marine copepods. Science 251: 794-796.

Reinfelder. J.R. and N.S. Fisher, N.S. (1994a) The assimilation of elements ingested by marine planktoic bivalve larvae. Limnol. Occ.tnog. 39: 12-20.

Reinfelder. J.R. and Fisher, N.S. (1994b) Retention o f elements adsorbed by juvenile fish (Menidia menidia. Menidia heryllina) from zooplankton prey. Limnol. Oceanog. 39: 1783-1789.

Rozas. L.P. and Lasalle, M.W. (1990) A comparison o f the diets of G ulf killifish. Fundulus grandis Baird and Girard, entering and leaving a Mississippi brackish marsh. Estuaries 13: 332-336.

Savvad. D. Stirling. A., and Toppin. G. (1975) Experimental studies on the effects o f copper on a marine food chain. Mar. Biol. 29: 351-359.

Smith. G.M. and Weis. J.S. (1997) Predator-prey relationships in mummichogs (Fundulus heteroclitus (L)): Effects o f living in a polluted environment.Journal o f Experimental Marine Biology and Ecology. 201: 75-87.

100

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sprague. J.B. (1986) Toxicity and tissue concentrations o f lead, zinc, and cadmium for marine molluscs and crustaceans. International Lead and Zinc Research Organization. Inc. Research Triangle Park, NC.

Stevenson. J.R. (1985) Dynamics o f the integument. InThe Biology o f Crustacea: Integument, Pigments, and Hormonal Processes (Eds. Bliss. D.E. and Mantle. L.H.). pp. 1-42. Academic Press, Orlando. FL.

Terwilliger. N.B. (1999) Hemolymph proteins and molting in crustaceans and insects. American Zoologist 39: 589-599.

Vandenhurk. P.. Faisal. M. and Roberst. M.H. (1998) Interaction o f cadmium and ben; [ajpyrene in mummichog (Fundulus heteroclitus): Effects on acute mortality. Marine Environmental Research. 46: 525-528.

Wallace. W.G. and Lopez. G.R. (1996) Relationship between subcellular cadmium distribution in prey and cadmium trophic transfer to a predator. Estuaries 19: 923-930.

Wallace. W.G. and Lopez. G.R. (1997) Bioavailability o f biologically sequestered cadmium and the implications of metal detoxification. Mar Ecol Prog Ser 147: 149-157.

Wang. W.X.. Reinfelder. J.R.. Lee. B.G .. and. Fisher. N.S. (1996) Assimilation and regeneration o f trace elements by marine copepods.Limnol. Ocecmog. 41: 70-81.

Wang. WX.. Qiu. J. W.. Qian. P.Y. (1999) Significance o f trophic transfer in predicting the high concentration o f zinc in barnacles. Enviro. Sci. Technol. 33: 2905-2909.

Wainright. S.C.. Weinstein. M.P.. Able. K.W. and Currin, C.A. (2000) Relative importance o f benthic microalgae, phytoplankton and the detritus o f smooth cordgrass Spartina alterniflora and the common reed Phragmites austrailis to brackish-marsh food webs. Mar Ecol Progr. Ser. 200:7-91.

Watras. C.J. and Bloom. N.S. (1992) Mercury and methylmercury in individual zooplankton: implications for bioaccumulation. Limnol. and Oceanog. 37: 1313-1318.

Weeks. J.M.. Rainbow. P.S.. and Moore. P.G. (1992) The loss, uptake and tissue distribution o f copper and zinc during the moult cycle in an ecological series o f talitrid amphipods (Crustacea: Amphipoda). Hydrohiologict 245: 15-25.

101

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. White, S.L., and Rainbow, P.S. (1984) Zinc flux in Palaemon elegans (Crustacea, Decapoda): molting, individual variation and tissue distribution. Mar. Ecol. Prog. Ser. 19: 153-166.

Zanders, I. P. and Rojas. W. E. (1996) Salinity effects on cadmium accumulation in various tissues o f the tropical fiddler crab Uca rapax. Environ. Pollut. 94: 293- 299.

Zhuang. Z. and Aheam. G.A. (1996) Calcium transport processes o f lobster hepatopancreatic brush border membrane vesicles. J. Exp. Biol. 199: 1195-1208.

Zyadah. M.A. and Abdel-Baky. I.E . (2000) Toxicity and bioaccumulation o f copper. zinc, and cadmium in some aquatic organisms. Bull. Environ. Contam. Toxicolo. 64: 740-747.

102

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LETTER OF PERMISSION

Kristen Keteles Louisiana State University 202 Life Sciences Building ELSEVIER Baton Rouge LA 70803 SCIENCE USA

ElaevterSdeact

Dear Kristen Keteles Global Righu PO Boa 100 MARINE POLLUTION BULLETIN, (in press), 2001, Keteles: "The contribution OifocdOXS IDX oftcdysis to the fate..." England

As per your letter dated 27 June 2001, we hereby grant you permission to reprintTtl: *U(0)ltd5S4JUO the Fax: *44 (0) ISAS 151)1} aforementioned material at no charge In your thesis subject to the followingE-mail: pcrm i>tiou#ilwv itr.co.uk conditions: Hone Page www.* luvitrxo m 1. If any part of the material to be used (for example, figures) has appeared in our publication with credit or aefaiowledgement to another source, permission must also be soughtfrom that source. If such permission is not obtained then that material may not be included in your publication/copies.

2. Suitable acbiowledgment to the source must be made, either as a footnote or in a reference list at the end of your publication, as follows:

“Reprinted from Journal title, Volume number, Authorfs), Title of article, Page Nos, Copyright (Year), with permission from Elsevier Science**.

3. Reproduction of this material is confined to the purpose for which permission is hereby given.

4. This permission is granted for non-exclusive world English rights only. For other languages please reapply separately for each one required. Permission excludes use in an electronic form. Should you have a specific electronic project in mind please reapply for permission.

3. This includes permission for UMI to supply single copies, on demand, of the complete thesis. Should his thesis be published commercially, please reapply for permission.

Yours sincerely

Frances Rothwfe^ (Mrs lobti Rights M

The processing of permission requests for a ll Elsevier Science (Including Pcrgamon Imprint) Journals has been centralised In Oxford, U K . Your future requests wffl be handled more qulekly I f yon w rite directly to: Subsidiary Rights Department, Elsevier Science, PO Box 800, Oxford OXS ID X , UK. Fax:44-1863 8533

Im p ru iu : Elsevier

103

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VITA

Kristen Annette Keteles was bom April11.1973 in Pittsburgh, Pennsylvania. She

graduated from Coastal Carolina University in 1995 completing a bachelor o f science

degree in marine science. She then worked as a research assistant through the Superfund

Basic Research Program at Dartmouth College on a project entitled. "Variation in

Bioaccumulation and Biomagnification o f Toxic Metals in Lakes o f the Northeastern

USA." She then joined the doctoral program at the Department o f Biological Sciences at

Louisiana State University in 1997. Presently, she is a candidate for the degree o f Doctor

o f Philosophy which w ill be awarded fall 2001.

104

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DOCTORAL EXAMINATION AND DISSERTATION REPORT

Candidates Kristen A. Keteles

Major Field: Zoology

Title of Diaaertation: Metal Partitioning among Tissues and Exoslceleton of Palaemonetes pugio and Its Role in Depuration and Trophic Transfer

A pproved:

jorf Profeaaor and Chairman

COMMITTEE:

Af)

Pate of Imagination:

July 13, 2001

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.